Molecular Phylogenetics and Evolution 61 (2011) 333–350

Contents lists available at ScienceDirect

Molecular Phylogenetics and Evolution

journal homepage: www.elsevier.com/locate/ympev

Phylogeography, geographic structure, genetic variation, and potential species boundaries in Philippine slender toads ⇑ Marites Bonachita Sanguila a,1, Cameron D. Siler b, , Arvin C. Diesmos c, Olga Nuñeza a, Rafe M. Brown b a Mindanao State University – Iligan Institute of Technology, Tibanga, Iligan City, Philippines b Biodiversity Institute and Department of Ecology and Evolutionary Biology, University of Kansas, Lawrence, KS 66045-7561, USA c Herpetology Section, Zoology Division, Philippine National Museum, Rizal Park, Burgos St., Manila, Philippines article info abstract

Article history: We investigated phylogeography of Philippine slender toads (genus Ansonia) and used a temporal frame- Received 22 December 2010 work for diversification, statistical tests of alternate topologies, and Bayesian approaches to test previous Revised 22 June 2011 hypotheses concerning dispersal to, and colonization routes within, the southern Philippine island of Accepted 23 June 2011 Mindanao. Two species of Ansonia previously have been documented, with ranges separated by an Available online 3 July 2011 east–west split corresponding to the approximate boundaries of Mindanao’s paleoisland precursors. We present new mtDNA sequence data (1946 bp from genes encoding ND1, 16S rRNA and tRNALeu) for Keywords: 105 Ansonia specimens sampled from 20 localities on Mindanao Island. Our data suggest that Philippine Conservation genetics Ansonia is composed of at least eight, well-supported population lineages, structured into a minimum of Dispersal Evolutionary Significant Units for four highly divergent mtDNA clades. One clade corresponds to Ansonia mcgregori, a range-restricted spe- Conservation (ESUs) cies apparently limited to the distal portion of the Zamboanga Peninsula of western Mindanao. Two mor- Torrent specialist larvae phologically indistinguishable, but genetically divergent, lineages possibly are undescribed cryptic Gene flow species from western Mindanao. We recognize the five remaining lineages as Ansonia muelleri pending Montane endemism data from morphology or bioacoustics that might diagnose separate species among these lineages. Regardless of their species status, the five allopatric lineages of A. muelleri should be viewed as important genetic units for future genetic conservation planning. Ó 2011 Elsevier Inc. All rights reserved.

1. Introduction tive or possibly outdated taxonomy (Alcala and Brown, 1998; Bickford et al., 2007; Brown, 2006; Brown and Alcala, 1994; Brown The geologically active islands of the Philippines possess high et al., 2000, 2008; Brown and Stuart, in press; Stuart et al., 2006). levels of endemic vertebrate biodiversity, which is predominantly With robust geographical sampling of genetic material from partitioned hierarchically into (1) Pleistocene Aggregate Island throughout the archipelago, a new group of studies have drasti- Complexes (PAICs; Inger, 1954; Heaney, 1985; Voris, 2000; Brown cally increased species diversity in several clades (e.g., Brown and Diesmos, 2002), (2) individual islands within PAICs, and (3) and Guttman, 2002; Brown et al., 2009; Evans et al., 2003a; montane subcenters of diversity within major landmasses (review: Linkem et al., 2010a, 2010b; Siler et al., 2010; Siler and Brown, Brown and Diesmos, 2009). This nested, highly partitioned nature 2010; Welton et al., 2010a, 2010b). Analyses of rates of species of the archipelago’s fauna has contributed to the recognition of description through time reveal rates of species discovery and the Philippines as a global conservation hotspot, with one of the description unparalleled in the history of Southeast Asian biodi- highest concentrations of land-vertebrate diversity on the planet versity studies (Brown and Diesmos, 2002; Brown et al., 2002, (Mittermeier et al., 1997, 1998, 1999; Reid, 1998; Brooks et al., 2008; Stuart and Bain, 2008; Brown and Stuart, in press; Siler 2002; Brown and Diesmos, 2009). However, the vast majority of et al., 2010, 2011). this diversity is based on species boundaries conceived by tradi- Relative to other parts of the archipelago, the large southern is- tional morphological taxonomy, calling attention to the possibility land of Mindanao has not received the same renewed research fo- of hidden or ‘‘cryptic’’ species diversity masquerading in conserva- cus, due principally to the inaccessibility of its many isolated mountains and logistical obstacles to field work. Much of what is known of the island’s high levels of herpetological diversity and ⇑ Corresponding author. Address: Biodiversity Research Center, 1345 Jayhawk endemism (Taylor, 1920, 1928; Inger, 1954; Brown and Alcala, Blvd., Lawrence, KS 66045, USA. Fax: +1 785 864 5335. 1970; Alcala and Brown, 1998; review: Brown et al., 2000, 2008) E-mail addresses: [email protected] (M.B. Sanguila), [email protected] comes from faunal inventories conducted during the early Euro- du (C.D. Siler). 1 Present address: Fr. Saturnino Urios University, San Francisco St., 8600 Butuan pean exploration (e.g., Boulenger, 1882, 1920; Van Kampen, City, Philippines. 1923; Smith, 1930, 1935), field work conducted in the early

1055-7903/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ympev.2011.06.019 334 M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350

1900s (Taylor, 1915, 1918, 1920, 1922a, 1922b), a single major warrant recognition as species or Evolutionary Significant Units expedition following World War II (Hoogstral, 1951; Inger, 1954), (ESUs) for conservation planning? (2) Are geographic patterns of and field work from the 1960s and 1970s (summarized in Leviton genetic variation consistent with stable, geographically structured, (1963), Brown and Alcala (1970, 1978, 1980)). Subsequent works populations or can we detect the signature of recent geographic or have been scattered, site-specific, and have not yet been synthe- demographic expansion? (3) Do genetic areas of endemism corre- sized in a biogeographic context (Smith, 1993a, 1993b; Amoroso, spond to the isolated montane areas of Mindanao, as would be pre- 2000; Delima et al., 2006, 2007; Nuñeza et al., 2010). Although en- dicted by natural history, larval morphology, and torrent-dwelling demic Mindanao species exemplars have been included in country- larval ecology (Inger, 1960; Brown and Alcala, 1982; Alcala and wide or regional phylogenetic studies (e.g., McGuire and Alcala, Brown, 1998; Inger, 1992)? (4) Can we reject Matsui et al.’s 2000; McGuire and Kiew, 2001; Brown and Guttman, 2002; Evans (2010) recent ‘‘early invasion’’ hypothesis as a general temporal et al., 2003a; Siler et al., 2011), no densely sampled phylogeograph- framework for the invasion of the southern Philippines and subse- ic or population genetic studies of Mindanao endemic vertebrate quent diversification of Philippine Ansonia? (5) Finally, can we re- groups have been conducted. ject hypothesized routes of dispersal along Mindanao’s elongate Here we present a phylogeographic study of the endemic Min- mountain chains as possible corridors for colonization and popula- danao slender toads of the genus Ansonia. Philippine species of tion expansions in order to account for the current distribution of Ansonia historically have been divided into two species, Ansonia Ansonia on the island of Mindanao? muelleri (Boulenger, 1887, from ‘‘Mindanao Island;’’ presumably eastern Mindanao; Inger, 1954, 1960; Alcala and Brown, 1998), 1.1. Geological setting and Mindanao biogeography and Ansonia mcgregori (Taylor, 1922c; from the southern tip of the Zamboanga Peninsula and nearby Basilan Island; Fig. 1). In Because the central portion of the island of Mindanao has been his first review of Philippine Amphibia, Inger (1954) questioned formed by a sequence of collision, accretion, and subduction events the validity of Taylor’s A. mcgregori. However, in later works he that have occurred over the past ten million years (Yumul et al., treated both A. mcgregori and A. muelleri as valid species (Inger, 2003, 2009; Hall, 1996, 1997), it is conceivable that the highly dy- 1960, 1966); other workers have not questioned this perspective namic history of the southern Philippines in part contributed to (Alcala and Brown, 1998; Brown, 2007). Finally, one recent spe- diversification of its fauna. An improved knowledge of the extent cies-level phylogenetic study (Matsui et al., 2010) supported the of land emergence (Lewis, 1997; Hall, 1998), combined with de- distinctiveness of two species of Ansonia on Mindanao and postu- tailed information concerning the timing of landmass collision lated an ancient (20 mya) invasion of the southern Philippines (Yumul et al., 2003; Hall, 1998), suggests possible dispersal routes for the pair of species on Mindanao. for fauna entering the southern portions of the Philippines in its In this paper we ask: (1) do robust genetic sampling and phy- early history. For example, although some components (particu- logeographic analyses support the recognition of only two species larly Zamboanga and extreme eastern Mindanao) may have been of Philippine Ansonia? Alternatively, do phylogeographic patterns land-positive greater than 15 mya, it is clear that they were very indicate the presence of additional divergent lineages that might far apart, and differed radically from today’s configuration (Hall, 1996, 1997, 1998). By 10–5 mya, the Sulu Archipelago–Mindanao arc was forming as a series of islands distributed west-to-east across the southern Philippines (Yumul et al., 2003, 2009; Hall, 1998, 2002). During this period, with extreme western and eastern Mindanao formed as widely separated islands (Hamilton, 1979; Hall, 2002), a pair of subduction zones, centered on the Cotabato Trench and the Philippine trench, respectively (Yumul et al., 2003), lead to the uplift of portions of central Mindanao. These up- lifts, combined with volcanic activity, produced many of the large mountains of the island (Hall, 1996; Yumul et al., 2008, 2009). To- day Mindanao’s isolated mountain ranges are each separated from others by wide, expansive, low-elevation plains and valleys. Thus, it is plausible that the temporally variable, dynamic emergence of the major upland montane regions of Mindanao first appeared as an island archipelago that spanned much of today’s central Min- danao. It is also conceivable that Ansonia and other taxa may have dispersed through what is today’s Mindanao by island hopping west-to-east across this paleoarchipelago. Many of the foothills of the mountains of central Mindanao are noted for low-elevation bands of marine sediments, suggesting past seashores, and the possibility that the major montane components likely existed as is- lands, formerly separated by shallow seas (Taylor, 1925, 1975; Hamilton, 1979; Hall, 1998). Thus, an hypothesis of an early inva- sion of the southern Philippines, followed by coincident or subse- quent diversification, may be at least partially consistent with Fig. 1. Map of Mindanao Island, southern Philippines (topographical relief indicated some of the available geological evidence (Yumul et al., 2009)if with elevational increments of increasingly dark shading) with 20 sampling localities indicated and approximate expected (on the basis of topographic relief the temporal framework for diversification was shown to be con- and larval biology of Ansonia; see Section 4) distribution of each genetic entity sistent with these earlier geological events. (haplotype clade, population lineage, or species) indicated with differently colored Biogeographers have identified the eastern Philippine island arc shading. Populations 1–3 correspond to A. mcgregori, populations 4 and 5 are as a possible dispersal route or entry point into the Philippines Ansonia sp. 1, population 6 is Ansonia sp. 2, and the remaining localities correspond (Diamond and Gilpin, 1983; Brown and Guttman, 2002; Brown to the five lineages of central and eastern Mindanao (Fig. 2), referred here to A. muelleri. (For interpretation of the references to color in this figure legend, the et al., 2009; Jones and Kennedy, 2008; Oliveros and Moyle, reader is referred to the web version of this article. 2010). This hypothesized route of biogeographic dispersion into M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350 335 the archipelago has been invoked in conjunction with an ‘‘island importantly, includes genetic material from each of the isolated hopping’’ mode of dispersal to explain the distribution of Minda- montane components of the island (Hall, 1996, 1997; Yumul nao faunal-region endemics (Brown and Guttman, 2002; Brown et al., 2004, 2008). Outgroup samples were selected based on the et al., 2009; Welton et al., 2010a, 2010b; Roberts et al., 2011). Sce- results of Matsui et al. (2010), which supported Ansonia guibei narios of this type typically assume a series of successive dispersal and A. fulginea as the closest relatives of Philippine lineages events via the Sulu Archipelago (the small chain of islands between (Appendix A); accordingly, we used Matsui et al.’s A. guibei Borneo and Mindanao; Fig. 1), Basilan, Mindanao, Dinagat/Siargao, sequence (Voucher KUHEL06B 054) and two sequences from Bohol, Leyte, and finally Samar islands (e.g., Brown and Guttman, A. fuliginea (BOR 22770 and KUHE 17537) as outgroups for this 2002; Brown et al., 2009). However, the curious absence of Ansonia study. on the northern islands of the Mindanao PAIC (e.g., Dinagat, Sia- For our 105 samples, we generated complete or partial se- rgao, Bohol, Leyte, or Samar; Brown and Alcala, 1970; Alcala and quences for the mitochondrial genes encoding ribosomal RNA Brown, 1998), plus only slight morphological character differences 16S (16S), NADH Dehydrogenase Subunit 1 (ND1), and a compo- between the two nominal species, suggests at least the possibility nent of a single transfer RNA (tRNAleu)(Appendix A). Only 16S se- of a recent (i.e., Pleistocene) invasion by a Bornean lineage as the quences were previously published and available as outgroups basis of distribution of Ansonia in the extreme southern Philip- (Matsui et al., 2010). pines. In a recent species-level phylogenetic analysis including a Genomic DNA was extracted from liver tissues stored in few samples from Mindanao, Matsui et al. (2010) demonstrated 95–100% ethanol following a guanidine thiocyanate protocol that the two Philippine endemics were each others’ closest rela- (Esselstyn et al., 2008). We used a combination of published and tives, moderately divergent (3–4% uncorrected sequence diver- newly developed primers to amplify targeted gene regions. Two gence) from one another and from Bornean congeners, and most primers were used to amplify a 840 bp region spanning most of closely related to A. fuliginea and A. guibei from Borneo (Inger, the 16S ribosomal RNA gene via polymerase chain reaction: 50 to 1960, 1966). That study suggested an ancient (20 mya) invasion 30: 16Sc GTRGGCCTAAAAGCAGCCAC and 16Sd CTCCGGTCTGAA of the southern Philippines by way of the Sulu archipelago (Matsui CTCAGATCACGTAG (Moriarty and Cannatella, 2004); PCR thermal et al., 2010). conditions followed Evans et al. (2003a). For ND1, we used ND1F2 CTACGTGATCTGAGTTCAGACCG and ND1R2 AAGGAGGTYC YTAWCTTTCGGGC primers and the same thermal profiles. 1.2. Larval biology, habitat, and ecological preferences of Ansonia Amplified products were visualized on 1.5% agarose gels. PCR products were purified with 1 lL of a 20% dilution of ExoSAP-IT The genus Ansonia contains 26–32 named and unnamed taxa (US78201, Amersham Biosciences, Piscataway, NJ) on the following (Matsui et al., 2010; Quah et al., 2011) distributed from Thailand thermal profile: 31 min at 37°, followed by 15 min at 80°. Cycle north of the Isthmus of Kra, through Peninsular Malaysia, and on sequencing reactions were run using ABI Prism BigDye Terminator the islands of Sumatra, Borneo (Indonesia and Malayisa), Basilan chemistry (Ver. 3.1; Applied Biosystems, Foster City, CA), and puri- and Mindanao (Philippines). An understanding of the larval ecol- fied with Sephadex (NC9406038, Amersham Biosciences, Piscata- ogy and morphology of toads of the genus Ansonia may provide way, NJ) in Centri-Sep 96 spin plates (CS-961, Princeton an explanation for their often patchy and topographically circum- Separations, Princeton, NJ). Purified products were analyzed with scribed pattern of montane endemism. Ansonia larvae are torrent an ABI Prism 3130xl Genetic Analyzer (Applied Biosystems). Con- specialists that possess a suite of highly derived morphological tinuous gene sequences were assembled and edited using Sequen- and ecological specializations (Inger, 1954, 1960, 1966, 1992, cher 4.8 (Gene Codes Corp., Ann Arbor, MI). All sequences were 1985, 2005; Duellman and Trueb, 1994; Matsui et al., 1998, deposited in GenBank (Accession Nos. JN314641-JN314839). 2005, 2007, 2009, 2010) enabling them exclusively to inhabit fast-flowing streams. Ansonia larvae possess specialized adhesive oral sucker mouthparts (Inger, 1985, 1992) allowing them to cling 2.2. Sequence data, alignment, geographic structure, and to rocks in the most rapid portions of high gradient, highly oxygen- phylogeographic analyses ated streams (Inger, 1954, 1966, 2005). Philippine Ansonia tadpoles are exclusively found adhering to rocks under white-water cas- Initial alignments were produced in Muscle (Edgar, 2004), and cades, around waterfalls, in shoots between boulders of steep, manual adjustments made in MacClade 4.08 (Maddison and Madd- rocky montane habitats, or in foothills or large mountains with ison, 2005). To measure phylogenetic congruence between the two topographic relief (RMB, personal observation). Unable to survive mitochondrial fragments, we inferred the phylogeny for each gene in slow-moving, low-gradient streams, Ansonia larvae and adults region independently using separate likelihood and Bayesian are thus elevationally restricted to montane slopes or low-eleva- methods. Following the observation of no incongruence between tion habits at the immediate base of mountains (where water cur- single-gene region topologies, we concatenated the data for subse- rents are sufficiently swift). Because of this unique larval quent analyses. Exploratory analyses of the combined dataset of morphology and natural history, we suspect that the low-lying val- 108 individuals (including 12 individuals with missing data for leys and inter-montane plains of Mindanao (Fig. 1) have been for- one of the two genes) and a reduced dataset of individuals with midable barriers to dispersal. Thus, past dispersal, population no missing data exhibited identical relationships. We therefore dispersion, and gene flow should have been necessarily restricted chose to include all available data for 108 individuals in subse- to the strips of suitable habitat along the slopes and foothills of quent analyses of the concatenated dataset. After excluding 98 bp major mountains of Mindanao. of ambiguous 16S rRNA and tRNA, the final dataset totaled 1946 aligned nucleotide positions. Our dataset was complete for 16S 2. Materials and methods and nearly complete for the ND1 region (missing ND1 only for the Bornean outgroups, two A. mcgregori samples, and seven A. 2.1. Taxon sampling and data collection muelleri samples). To assess general patterns of genetic diversity within clades, we In-group sampling included 105 individuals collected from 20 calculated the numbers of haplotypes (N), haplotype diversity (h; localities throughout Mindanao Island (Fig. 1; Appendix A). Our Nei, 1987), numbers of polymorphic sites, and nucleotide diversity sampling maximizes geographical coverage of the island and, (p; Nei and Tajima, 1981) using DNASP 4.0 (Rozas et al., 2003) and 336 M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350

Arlequin 3.1 (Excoffier et al., 2005) for each major lineage of Min- larval biology and specialization to torrent habitats has restricted danao Ansonia following the methods of Brown et al. (2010). dispersal of Mindanao Ansonia to mountain slopes and the foothills Parsimony analyses were conducted in PAUP⁄ 4.0b 10 of Mindanao’s major montane components. We generated hypoth- (Swofford, 2002), with gaps treated as missing data and all charac- eses of phylogeographic history based primarily on the geological ters weighted equally. Most-parsimonious trees were estimated history of Mindanao Island (Hamilton, 1979; Yumul et al., 2003, using heuristic searches with 1000 random addition-sequence rep- 2004, 2008, 2009; Hall, 1996, 1997, 1998, 2002), and the general licates and tree bisection and reconnection (TBR) branch swapping. expectations that (1) Ansonia populations should be confined to To assess clade confidence, nonparametric bootstrapping was con- strips of suitable habitat along Mindanao’s elongate mountain ducted using 1000 bootstrap replicates, each with 100 random chains and (2) an overall patterns of west-to-east invasion and col- addition-sequence replicates and TBR branch swapping. onization should have occurred (and, in fact, was detected in the Partitioned Bayesian analyses were conducted in MrBayes v3.1.2 signal of a previous species-level phylogeny for the genus; Matsui (Ronquist and Huelsenbeck, 2003). ND1 was partitioned by codon po- et al., 2010). The topological constraints that we formulated to test sition; 16S and the single tRNA (tRNAleu) were treated as a single par- hypothesized routes of dispersal are illustrated in Fig. 3. tition. The Akaike Information Criterion (AIC), as implemented in We evaluated hypotheses with two methods, via a Bayesian ap- jModeltest v0.1.1 (Guindon and Gascuel, 2003; Posada, in press), proach of evaluating alternative topologies contained in post-bur- was used to select the best-fit model of nucleotide substitution for nin samples and with Approximately Unbiased tests (AU; each partition (Table 1). The best-fit model for each data partition Shimodaira and Hasegawa, 2001; Shimodaira, 2002). was implemented in subsequent Bayesian analyses. A rate multiplier We began by constructing constraint trees consistent with our a model was used to allow substitution rates to vary among subsets, priori hypotheses (see below) in McClade. Post-burnin trees were and default priors were used for all model parameters. We ran four then filtered (in PAUP) for topologies consistent with these con- independent Metropolis-coupled MCMC analyses, each with four straint topologies. Constraint topologies were conservatively con- chains and an incremental heating temperature of 0.05. All analyses structed to address only specific components of each dispersal were run for 40 million generations, sampling every 5000 genera- scenario (Fig. 3). For the Bayesian implementation, we took the tions. To assess stationarity, all sampled parameter values and log- percentage of 6400 post-burning trees consistent with individual likelihood scores from the cold Markov chain were plotted against constraint hypotheses to represent the posterior probability that generation time and compared among independent runs using Tracer explains our data. v1.4 (Rambaut and Drummond, 2007). Furthermore, we plotted the AU tests were then performed on the per-site likelihoods using cumulative and non-overlapping split frequencies of the 20 most var- CONSEL v0.1i (Shimodaira and Hasegawa, 2001), following the iable nodes, and compared split frequencies among independent runs methods of Siler et al. (2010). The p-value reported for a given using Are We There Yet? [AWTY (Wilgenbusch et al., 2004)]. Although hypothesis is the largest p-value of all the trees inferred under that all samples showed patterns consistent with stationarity after 2.5 constraint. We automated the process with Perl and Python scripts million generations (i.e., the first 12.5%), we conservatively discarded (written by J. Oaks and CDS; available by request). the first 20% of samples as burn-in. The first hypothesis that we tested addresses the expectation of Partitioned maximum likelihood (ML) analyses were conducted a west-to-east, stepping-stone mode of dispersal up the Zambo- in RAxMLHPC v7.0 (Stamatakis, 2006) on the concatenated dataset anga Peninsula following presumed dispersal from Basilan Island using the same partitioning strategy as above. The complex (Matsui et al., 2010). For this hypothesis (Fig. 3a) we constructed GTR + C model was used for all subsets, and 100 replicate ML infer- a grade-like constraint topology, with western populations con- ences were performed. Each inference was initiated with a random strained to be the sister taxon to populations from other areas of starting tree, and employed the rapid hill-climbing algorithm (Sta- modern Mindanao. This hypothesis was very similar to the taxo- matakis et al., 2007). Clade confidence was assessed with 100 boot- nomic hypothesis (A. mcgregori + A. muelleri), above, except that strap pseudoreplicates (Stamatakis et al., 2008). Lake Lanao populations from extreme western Mindanao were constrained to be the sister taxon to the remaining populations of Mindanao (arranged in a polytomy). For Hypothesis 2 (Fig. 3b), 2.3. Hypothesis testing: taxonomy and dispersal corridors we left Zamboanga and western Mindanao populations free to vary outside the clade comprising 7–15, and tested a hypothesized Current taxonomy suggests the occurrence of one species (A. route of dispersal from west-central highlands to east-central muelleri) in central and eastern Mindanao (Boulenger, 1887; Tay- highlands, south via the Apo Massif, with final dispersal events into lor, 1922a, 1922b, 1922c; Inger, 1954, 1960, 1966; Alcala and the southern mountains of Cotabato and Sarangani Provinces. In Brown, 1998; Brown, 2007) and another (A. mcgregori) confined Hypothesis 2, we did not constrain eastern Mindanao populations to the extreme western Mindanao Zamboanga Peninsula. We (16–20) in any way. For Hypothesis 3 (Fig. 3c), we tested an alter- tested this hypothesis by searching the post burn-in sample for native route of dispersal into central Mindanao, namely the south- topologies consistent with a simple east–west geographic split, ern route along the Cotabato coast, the Mt. Busa Range, the Apo corresponding to the suture (Fig. 1) between the Zamboanga Pen- Massif, and finally the east-central highlands. As before, we did insula and the remaining portions of Mindanao. not constrain branches leading to western or eastern Mindanao Additionally, we evaluated five hypotheses of dispersal routes populations, but allowed these terminals to vary freely outside and phylogeographic history (Fig. 3) based on the assumption that the clade comprising populations 6–8 + 11–15 in analyses. In Table 1 Hypotheses 4 and 5, we addressed the origins of the eastern Models of evolution selected by AIC and applied for partitioned, model-based Mindanao populations. For Hypothesis 4 (Fig. 3d) we tested for evi- phylogenetic analyses. dence of a northern entryway into eastern Mindanao (east-central Partition AIC Model Number of highlands, northeast mountains, southeast mountains), and for model applied characters Hypothesis 5 (Fig. 3e) we constructed a constraint representing ND1, 1st codon position TVM + C GTR + C 310 the opposite route, namely dispersal into eastern Mindanao via ND1, 2nd codon position F81 HKY 310 the Apo Massif, to the southeastern mountains, and finally the ND1, 3rd codon position TrN + C GTR + C 309 northeastern range. As in the case of the previous taxonomic 16S TVM + C GTR + C 944 hypothesis of two species, we used the percentage of 6400 tRNALeu HKY HKY 73 post-burnin trees that were consistent with each hypothesis as M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350 337

Fig. 2. Preferred phylogenetic estimate inferred from Bayesian, Likelihood, and Parsimony (not shown) analyses of mitochondrial DNA sequence data. Nodes supported by P95% Bayesian PP and ML bootstrap support were considered significantly supported and are indicated by black circles. Terminals are labeled with abbreviated site names, followed by general geographic distribution (Appendix A). the posterior probability that an individual hypothesis explained qualitatively assessed mismatch distributions for multimodal our data. (which could stem from a structured population) versus smooth unimodal (which may be indicative of possible recent population 2.4. Inference of historical demography expansion or sudden panmixia) distributions and calculated rag- gedness indexes and their significance in Arlequin (Rogers and Because of the uncertainty with regard to taxonomy of Minda- Harpending, 1992; Harpending et al., 1998). We also calculated nao Ansonia (see Sections 3 and 4), wherever possible, we assessed Tajima’s D (implemented as a test for selective neutrality), and Ra- each of the various lineages of Mindanao Ansonia for evidence of mos and Rozas R2 statistics (Ramos-Onsins and Rozas, 2002), as recent change in effective population size in nested fashion. First, additional indicators of potential population expansion. Finally, we performed population genetic and historical demographic anal- we investigated the possibility of population or demographic yses at a fine scale, focusing separately on six of the eight (those expansion using Fu’s Fs neutrality test (Fu, 1997). Fu’s F assumes with sufficient population sampling) empirically observed haplo- neutrality and may diagnose recent population or demographic type groups. These included one haplotype clade clearly assignable expansion via a highly negative value of Fs. to A. mcgregori (see below) and also each of the five divergent and Because Fs and R2 are summary statistics (based on distribu- allopatric haplotype clades of central and eastern Mindanao (see tions of haplotypes and numbers of segregating sites) they do Section 4). We then combined these five minimally- to moder- not use all of the historical information contained in DNA sequence ately-divergent lineages into a single analysis to examine popula- variation (Galbreath et al., 2009). Alternatively, we assessed tion genetic and demographic parameters under a conservative changes in demographic expansion of estimated effective popula- hypothesis of a single, widespread species (putatively A. muelleri). tion size over the history of each major lineage by applying Bayes- We calculated mismatch distributions in Arlequin 3.1 (Excoffier ian skyline procedures (Drummond et al., 2005) in BEAST 1.6.1 et al., 2005), which determines significance via coalescent simula- (Drummond and Rambaut, 2006, 2007) to the clade comprising tions of a large, neutrally evolving population of constant size in A. mcgregori, three of Mindanao’s five central and eastern gene lin- the context of assumed selectively neutral nucleotide substitutions eages separately (those with sufficient sampling: Busa–Kiamba, (Slatkin and Hudson, 1992; Rogers and Harpending, 1992). We Balatukan–Tago–Calabugao, and Eastern Mindanao), and a final 338 M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350

Fig. 3. Five historical hypotheses (constraint topologies) employed to test dispersal routes (montane corridors) to account for the observed pattern of occurrence of Ansonia lineages on Mindanao Island. Numbered populations on tree terminals correspond to sampling localities in Fig. 1. See text for detailed explanation of each hypothesis. PP = Posterior Probability; AU = p-value of each hypothesis as inferred from the Approximately Unbiased tests (AU; Shimodaira and Hasegawa, 2001; Shimodaira, 2002). analysis with these five lineages combined (A. muelleri), as de- invasion of Mindanao. If early invaders of the Mindanao paleoar- scribed above. We attempted to diagnose fluctuations in popula- chipelago were widespread when these landmasses collided tion median effective size over time for each population by (10–5 mya) and the major period of mountain-building began with approximating the posterior distribution of effective population subduction at the Cotobato and Philippine trenches (Yumul et al., size (from the original, unreduced dataset) over intervals of the 2003, 2008, 2009), diversification of Philippine Ansonia could have phylogeny. For these two analyses we chose the appropriate model been associated with orogenic vicariance during the uplift and ele- of sequence evolution with AIC in jModeltest, and generated input vational isolation of Mindanao’s major mountain chains (Hall, files with BEAUTi, using the same or closest available (and next- 1996; Yumul et al., 2003). We evaluate this possible scenario in most, parameter-rich) model. We ran both analyses for 20 million the context of our temporal framework provided by divergence- steps, using default parameters, sampling every 1000 steps, and date estimation. conservatively discarded the first 10% of samples as burn-in. We As a test of the Matsui et al. (2010) calibration, we employed an employed a fixed substitution rate of 1.4%/my, which was inferred alternative approach and inferred approximate divergence dates from the entire dataset and the placement of an uncorrelated log- with a Bayesian relaxed molecular clock in BEAST v1.6.1. A variety normal prior on a mean substitution rate spanning a range of sub- of recent empirical studies of amphibians (using a variety of cali- stitution rates inferred from recent studies involving amphibian brations, taxa, and mtDNA gene fragments) have inferred model- mitochondrial DNA (see below). corrected mitochondrial sequence divergence rates of between Analyses were independently replicated four times with differ- 0.8 and 1.9% total divergence per million years (Tan and Wake, ent random starting seeds, and results then were combined in 1995; Macey et al., 1998, 2001; Crawford, 2003a, 2003b; Wang LOGCOMBINER 1.4.7 (after burn-in) after examining convergence et al., 2008) for various parts of the genome. We used this range diagnostics in TRACER 1.4 (Rambaut and Drummond, 2007). and a relaxed clock approach as implemented in BEAST, allowing branch lengths to vary according to an uncorrelated lognormal dis- 2.5. Timing of diversification tribution, employing a Yule process tree prior, with all remaining priors set to default. We selected a rate prior with a normal distri- The absence of a fossil record for Ansonia or closely related bution and a mean of 1.4% (with a CI of 0.008–0.019) and simulta- members of the family Bufonidae renders precise and accurate neously estimated divergence dates and chronogram topology. For estimation of divergence times difficult. Matsui et al. (2010) ex- this analysis we selected the GTR + C model and ran four indepen- tended their analysis of species relationships in Ansonia by inclu- dent analyses of 60,000,000 steps each, reviewed convergence sion of outgroup bufonids from the Pramuk et al. (2007) dataset, diagnostics in Tracer v1.5 (Rambaut and Drummond, 2007), and permitting two external calibration points for a simultaneous combined runs in LogCombiner v1.6.1 after confirming conver- Bayesian estimation of phylogeny and ingroup divergence times. gence and conservatively excluding 10,000 trees as burn-in. Final- That approach yielded an estimated mean age of 20.2 MY ly, we summarized results as a maximum clade credibility tree (9.8–31.8 HPD) for the node subtending A. mcgregori + A. muelleri (Drummond and Rambaut, 2006, 2007) in TreeAnnotater v.1.6.1. (Matsui et al., 2010). We recognize that any approach lacking multiple reliable inter- The ancient above-water existence of the Zamboanga Peninsula nal calibration points is subject to a variety of sources of error and its proximity to the Sunda Shelf makes plausible a very early (Graur and Martin, 2004). However we proceeded with the analysis M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350 339 presented here to provide an approximate estimate for diversifica- phylogenetic signal and a lack of any conflict among optimality tion of Mindanao Ansonia and to provide an alternative to past criteria. analyses (Matsui et al., 2010), which we suspect presented All analyses suggest the existence of eight markedly divergent unrealistically old dates for some of the same nodes in our tree. population lineages diagnosed by mitochondrial haplotype clades Additionally, this approach allows us to consider the speciation- (Figs. 1 and 2). Each of these eight haplotype clades is strongly sup- by-orogenic-vicariance scenario, described above. ported with likelihood bootstrap values of P95 and posterior prob- abilities of P0.95 (Fig. 2). With one exception, relationships among these eight clades are also strongly supported (MLBP P 95; 3. Results PP P 0.95). The exception involved moderate likelihood support (MLBP = 71) and Bayesian support (PP = 0.80) for the placement 3.1. Taxon sampling, sequence data, and alignment of the Rogongon and Kitanglad lineage as the sister taxon to the remaining samples referred to A. muelleri in our preferred phyloge- The complete, aligned matrix contains 105 newly sequenced netic hypothesis (Fig. 2). Mindanao’s eight divergent genetic lin- samples of Mindanao Island slender toads. Three additional sam- eages of Ansonia exhibit pronounced geographical structure ples were included as outgroups, consisting of representatives of (Fig. 2) and are clearly associated with the major upland montane Ansonia guibei and A. fuliginea. Following initial unrooted analyses, components of this island (Fig. 1). and the results of a recent phylogenetic study of Ansonia (Matsui Specimens referable to A. mcgregori were represented by 16 et al., 2010), we rooted the tree using samples of A. guibei and A. unique haplotypes from numerous specimens from the type fuliginea. Within each gene region, variable and parsimony-infor- locality at the tip of the Zamboanga Peninsula (Nancy + Baluno; mative characters were observed as follows: 174 and 147 out of Pasonanca Natural Park, Zamboanga City area), and samples 944 (16S); 166 and 155 out of 929 (ND1); 10 and 10 out of 73 from Murias (a locality located midway along the Zamboanga (tRNAleu). Peninsula). The remaining central and eastern Mindanao sam- pling consists of five divergent lineages, representing 42 unique 3.2. Phylogeography and analyses of population structure haplotypes, which we tentatively assign to A. muelleri (see Sec- tion 4). These geographically structured clades are moderately We identified 58 unique haplotypes among the Philippines divergent (separated by uncorrected p-distances of 1.6–3.8%) Ansonia (Table 2). These form four major clades, corresponding to and each appears restricted to geographically distinct, upland A. mcgregori (including the type locality; sixteen distinct haplo- areas of Mindanao (Fig. 1). Within the five clades putatively as- types and 25 polymorphic sites), a clade from Murias and Timolan signed to A. muelleri, a strongly supported (MLBP P 95; (4 haplotypes and 33 polymorphic sites), a clade represented by PP P 0.95) eastern and eastern clade of Magdiwata, Compostela, two samples from Lake Lanao (2 haplotypes and 3 polymorphic Hamiguitan, Surigao del Norte, and Agusan del Norte populations sites), and five divergent mitochondrial gene lineages, forming a is the sister taxon to a divergent lineage from southern Minda- clade we putatively assign to A. muelleri (see Section 4; 42 haplo- nao (Kiamba and Mt. Busa populations; Figs. 1 and 2). These types with 224 polymorphic sites). Within these four clades, mean two clades are moderately supported (MLPB < 70; PP = 0.70) as number of pairwise nucleotide differences (k), haplotype diversity most closely related to a clade from east-central Mindanao (h), and percent nucleotide diversity (p) are all relatively low (Balatukan, Tago, and Calabugao populations), and the weakly (Table 2). As expected, pairwise nucleotide differences (k) and supported cluster containing these three clades is strongly sup- nucleotide diversity (p) were highest in the large geographic ported as most closely related to specimens from the Mt. Apo sample referred to A. muelleri (especially the Rogongon, Balatukan, Massif (Figs. 1 and 2). The Rogongon and Kitanglad clade is only and Eastern Mindanao lineages; Table 2), but interestingly k and p moderately supported (MLBP 71; PP = 0.80) as the sister clade to were also relatively high in the Murias/Timolan lineage, despite the the remaining clades referred to A. muelleri (Fig. 2). fact that only four individuals were sequenced in this group. All A lineage represented by two unvouchered genetic samples measures of sequence diversity were lowest in A. mcgregori and from the Lake Lanao (Marawi) area is the sister taxon to A. mcgre- in the Lake Lanao clade (Table 2). gori + A. muelleri and separated from samples of these two species Phylogenetic analyses of the combined ND1 + 16S data yielded by 3.8–5.0% uncorrected sequence divergence (Table 3). A second well-resolved topologies with high bootstrap support (MP and highly divergent lineage from Murias and Timolan is the sister tax- ML) and posterior probabilities throughout most trees (Fig. 2; on to the (Lake Lanao (A. mcgregori + A. muelleri)) clade and sepa- MP results not shown). Topologies were not statistically incon- rated from these lineages by 6.2–7.8% uncorrected sequence gruent regardless of the analytical method, suggesting strong divergence (Table 3).

Table 2

Summary of southern Philippine Ansonia sampling, major lineages/haplotype clades, numbers of individuals (N), numbers of mtDNA haplotypes (Nh), numbers of polymorphic sites (PN), mean number of pairwise nucleotide differences (k), haplotype diversity (h), and nucleotide diversity (p). See Appendix A for full details of sampling and a list of all samples included.

Region/Clade NNh PN khp Ansonia sp. 1 4 4 33 16.667 ± 9.46 1.000 ± 0.18 0.0086 ± 0.0058 Ansonia sp. 2 2 2 3 3.000 ± 2.45 1.000 ± 0.50 0.0037 ± 0.0042 A. mcgregori 19 16 25 4.614 ± 2.37 0.983 ± 0.02 0.0024 ± 0.0014 Rogongon–Kitanglad 7 6 8 3.333 ± 1.98 0.933 ± 0.12 0.0017 ± 0.0012 Apo 8 4 1 0.500 ± 0.52 0.500 ± 0.27 0.0003 ± 0.0003 Balatukan–Tago 26 13 40 9.822 ± 4.65 0.89 ± 0.04 0.0050 ± 0.0027 Kiamba–Busa 17 7 23 4.542 ± 2.36 0.68 ± 0.11 0.0023 ± 0.0014 Eastern Mindanao 22 12 82 22.808 ± 10.60 1.00 ± 0.02 0.0117 ± 0.0061 All A. muelleri samples 80 42 224 47.586 ± 20.86 0.966 ± 0.01 0.0245 ± 0.0119 All samples 105 64 1269 110.671 ± 47.93 0.981 ± 0.01 0.0569 ± 0.0273 340 M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350

Table 3 Uncorrected p-distances (16S + NAD1 gene sequence divergence, expressed as percentages) among and within mitochondrial gene lineages of Mindanao river toads, genus Ansonia. Percentages on the diagonal represent within-clade genetic diversity (bolded for emphasis).

A.sp1 A.sp2 A. mcgregori Rogongon Apo Balatukan Kiamba E. Mindanao 1 0.1–0.3 2 5.5–6.0 0.0 3 6.9–7.3 3.7–4.0 0.0–0.3 4 6.3–6.7 4.1–4.3 3.0–3.3 0.0–0.2 5 6.2–7.8 4.3–4.4 2.9–3.9 2.7–3.4 0.0–0.1 6 7.4–7.7 3.8–4.1 3.4–3.8 2.7–2.9 1.6–2.8 0.0–0.8 7 7.1–7.7 4.4–5.0 4.0–4.4 3.2–3.6 2.3–3.4 2.6–3.0 0.0–0.6 8 6.1–7.8 4.3–4.5 3.0–4.3 2.7–3.8 1.9–3.8 2.1–3.1 2.3–3.5 0.0–1.4

3.3. Hypotheses tested: previous taxonomy and dispersal corridors served value statistically significant (p < 0.05), consistent with the hypothesis of population growth in both species. Our data reject the hypothesis that Philippine Ansonia consists Bayesian skyline plots revealed qualitatively different demo- of only two species, separated geographically by the suture be- graphic histories when considering the three central and eastern tween the Zamboanga Peninsula and the remaining portions of Mindanao clades (where samples sizes were sufficient), or when Mindanao. The hypothesis that all Zamboanga populations these lineages were pooled and assumed to represent A. muelleri (A. mcgregori) are the sister taxon to all others (A. muelleri) was re- (Fig. 5). When individual skyline analyses were conducted on the jected by both Bayesian (PP = 0.0) and likelihood-based methods three central and eastern Mindanao gene lineages for which sam- (AU p-value < 0.001). Our Bayesian tests of dispersal via montane pling was sufficient, each one revealed slight demographic or range corridors strongly reject the simple east-to-west dispersion contraction, followed by expansion after a longer period of demo- hypothesis and all other hypothesized dispersal routes, with pos- graphic stability. When we pooled data and comparaed putative terior probabilities indicating a complete absence of any of the a species (Fig. 5), A. mcgregori appears to have experienced gradual priori topologies in our post burn-in sample of trees. However, population expansion over the past 1.0–0.7 my, but A. muelleri the AU test indicated a significant but marginal (AU = 0.01) rejec- may have undergone more rapid and recent demographic or range tion of the ‘‘southern route’’ (Fig. 3c) hypothesis, and failed to re- contraction (0.3 mya) following a period of relatively stable ject (AU = 0.244) the ‘‘east-central to northeastern’’ hypothesized demographic history. dispersal route (Fig. 3d).

3.5. Timing of diversification 3.4. Historical demography As expected, our estimate of the temporal framework for diver- With the caveat that our sample sizes vary, mismatch distribu- sification within Mindanao Ansonia provided much younger dates tions showed striking differences between the two described spe- than those postulated by Matsui et al. (2010) for the approximate cies (Fig. 4). When mismatch distributions and associated statistics root age and oldest divergences within this clade (Fig. 6). Most were calculated individually for the five divergent gene lineages of divergences associated with recognized and hypothesized species central and eastern Mindanao, these all appeared qualitatively rag- boundaries appear to have occurred well within the Pleistocene ged, with the exceptions of the Rogongon and Apo haplotype (<1.8 mya; Fig. 6). The highest posterior density intervals around clades in which samples sizes were low (Fig. 4). Given the lack of our estimates are admittedly wide, but even with substantial statistically significant differences between these individual gene uncertainty our analysis places the estimated root age of Mindanao lineages (excepting the Rogongon and Eastern Mindanao clades Ansonia at approximately 2.3 mya (HPD = 1.19–3.40 mya), far where Fu’s Fs was significant; Table 2) further discussion will focus younger (an order of magnitude; HPD intervals non-overlapping) on the combined analysis in which we pooled these data and puta- than 20.2 (HPD = 9.8–31.8 mya) estimated by Matsui et al. tively consider all of these lineages to be members of a single spe- (2010). Our estimates suggest that A. mcgregori diverged cies, A. muelleri. Whereas sequence data of A. mcgregori exhibited a 0.87 mya (HPD = 0.42–1.53 mya) and that the node subtending smooth, unimodal distribution of pairwise differences suggesting a the five lineages that we conservatively assigned to A. muelleri orig- lack of geographical structure or a population expansion event, the inated 1.77 (HPD = 0.99–2.78) mya. Most of the geographically five gene lineages we refer to A. muelleri were characterized by a structured clades (Fig. 6) include mean estimated divergences dur- relatively ragged and multimodal distribution, suggesting struc- ing a period of approximately 1.0–0.5 mya, suggesting that geo- tured, demographically stable populations. The T statistic and Rog- graphic structure was in place well before the last glacial ers and Harpending’s Raggedness Index are nonsignificant for A. maximum (12–10 kya). We consider these findings to refute both mcgregori but significant for A. muelleri (Table 4), indicating a fail- an ‘‘early-invasion’’ hypothesis (20 mya; Matsui et al., 2010) ure to reject the sudden expansion model in A. mcgregori but a and a ‘‘speciation-by-orogenic-vicariance’’ scenario (10–5 mya). demographically stable population inferred in A. muelleri. Tajima’s D indicated no departures from expectations of neutrality and demographic stability for either lineage (p > 0.05; Table 4). Fu’s 4. Discussion Fs is positive, close to zero, and nonsignificant in A. muelleri (con- sistent with an hypothesis of constant population size), but large 4.1. Taxonomic implications in magnitude, negative, and significant (p < 0.001) in A. mcgregori, rejecting the hypothesis of demographic stability and consistent How many species of Mindanao Island slender toads are im- with an interpretation of recent range expansion. Ramos-Onsins plied by our data? First, we collected the morphologically distinct and Rozas R2 statistics, said to be a superior indication of range A. mcgregori at the species’ type locality (Pasonanca Natural Park, at or demographic expansion in small populations (Ramos-Onsins the western tip of the Zamboanga Peninsula; Taylor, 1922c)sowe and Rozas, 2002), were small and positive (Table 4), and the ob- can be certain of this species’ identity; our data strongly support its M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350 341

Fig. 4. Observed frequencies of pairwise nucleotide differences among sequences (see key; black lines) and expected frequencies under a model of sudden population expansion (gray lines) (Rogers and Harpending, 1992) for all populations referred to Ansonia muelleri, each of the five clades of A. muelleri from central and eastern Mindanao (see Figs. 1 and 2: Rogongon–Kitanglad, Apo, Balatukan-Tago-Calabugao, Kiamba-Busa, and eastern Mindanao), and A. mcgregori. The relatively smooth and unimodal plot for A. mcgregori matches predicitons for a demographically expanding population; the ragged and multimodal plots for A. muelleri indicate long-term demographic stability.

Table 4 Summary statistics and results of tests of population expansion in Philippine Ansonia (A. mcgregori and the gene lineages referred to A. muelleri; divergent lineages A. sp. 1 and A. sp. 2 excluded): analysis of mismatch distributions, and substitution model applied to the Bayesian skyline analysis of demographic history. For mismatch distributions, T is presented along with p-values for rejection of the sudden expansion model, based on a comparison of the sum of squares of expected and observed distributions (using parametric bootstrapping with 10,000 replicates; Rogers and Harpending, 1992; Excoffier et al., 2005). Additional entries include Harpending’s Raggedness Index (RI) and p-values for rejection of the goodness of fit test comparing simulated vs. observed distribution raggedness, Tajima’s D, Fu’s Fs, and Ramos-Onsins and Rozas R2 statistics. All tests were implemented separately for the two species A. mcgregori and A. muelleri as defined in the text.

a b 2 Species TRITajima’s D (p-value) Fs(p-value) R2 (p-value) Skyline Model A. mcgregori (19/16) 3.954 (0.930) 0.016 (0.889) 1.396 (0.081) 10.953 (0.000) 0.137 (0.019) HKY Rogongon-Kitanglad (7/6) 1.680 (0.510) 0.129 (0.620) 0.185 (0.477) 3.174 (0.001) 0.052 (0.341) HKY Apo (8/4) 0.768 (0.320) 0.250 (0.890) 0.612 (0.414) 0.172 (0.339) 0.052 (0.454) HKY Balatukan–Tago (26/13) 17.438 (0.460) 0.022 (0.640) 0.897 (0.194) 0.164 (0.496) 0.101 (0.111) GTR Kiamba–Busa (17/7) 6.134 (0.150) 0.209 (0.140) 0.740 (0.251) 0.778 (0.691) 0.039 (0.09) HKY Eastern Mindanao (22/12) 27.166 (0.780) 0.021 (0.540) 0.178 (0.449) 5.6178 (0.012) 0.051 (0.01) GTR All A. muelleri (80/42) 59.229 (0.040) 0.009 (0.000) 0.385 (0.342) 0.013 (0.590) 0.162 (<0.001) GTR + C

a Numbers of individuals/numbers of haplotypes in parentheses. b Statistical significance for rejecting a null model of constant population size included in parentheses. 342 M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350

Fig. 5. Bayesian skyline plots for A. mcgregori, the three clades of A. muelleri from central and eastern Mindanao with sufficient numbers of samples (see Figs. 1 and 2: Kiamba- Busa, Balatukan-Tago-Calabugao, and eastern Mindanao), and for all populations referred to A. muelleri. Bold black lines indicate an estimate of median effective population size as a function of time; gray lines indicate the 95% highest posterior density. The horizontal axis has been scaled to show the same time interval both plots, resulting in the truncation of the A. mcgregori trace. The vertical axis should be interpreted as the change in effective population size. status as a distinct species in confirmation of earlier studies (Tay- eages as ‘‘Ansonia sp. 1.’’ and ‘‘Ansonia sp. 2’’ (Fig. 3; Table 3)to lor, 1922c; Inger, 1966). emphasize their genetic distinctiveness and to highlight the need Our phylogenetic estimates and analyses of sequence diver- for additional study. gence suggest two highly divergent haplotype clades (Fig. 3; Ta- Given no strong statistical support for individual demographic, ble 3), one found at both Murias and Timolan and another at Lake population genetic, or divergent phylogenetic histories in the cen- Lanao (Fig. 1). Our preliminary survey of available specimens tral and eastern haplotype mtDNA gene lineages, we tentatively (Appendix A), plus the types of both A. mcgregori and A. muelleri, identify the widespread, highly structured clade from central and identified no character differences of external morphology that eastern Mindanao as a single species (A. muelleri). This arrangement would diagnose these two putative evolutionary lineages as new includes five geographically structured haplotype clades, separated species (Sanguila, Fuiten, and Brown, unpublished data). At present, from one another by minimal to moderate (1.6–3.8%) sequence due to the potential for species boundaries and relationships to be divergence (uncorrected p-distance; Table 3). Although the exact obscured by mitochondrial lineage sorting and introgression, main- type locality for A. muelleri is not known (the species was described tenance of ancestral polymorphisms, and deep coalescent events by Boulenger, 1887, from ‘‘Mindanao Island’’), and the holotype (Hudson, 1990; Nichols, 2001; Hare, 2001), it is premature to recog- specimen (BM 1947.2.20.57) falls within the range in morphologi- nize species without diagnostic characters derived from an inde- cal variation we have observed (Sanguila, Fuiten, and Brown, pendent source of data (e.g., adult or larval morphology, unpublished data), we follow many other herpetologists in assum- bioacoustics, etc.). We refer to these highly divergent genetic lin- ing that A. muelleri was originally collected in central or eastern M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350 343

Fig. 6. Bayesian chronogram inferred by using a relaxed molecular clock as implemented in BEAST v1.6.1. Gray nodal bars are highest posterior density intervals, and the time axis is indicated at the bottom (mya). Nodal support values indicate Bayesian posterior probabilities of >0.50 (asterisks = posterior probabilities of >0.99). Note that most major divergences occurred within the last 1.8 my, well within the Pleistocene Epoch.

Mindanao (Inger, 1960, 1965, 1966; Brown and Alcala, 1970; Alcala, Our results, together with a new body of well-sampled, geo- 1986; Brown and Alcala, 1989; Brown, 2007). The hypothesis of a graphically robust phylogenetic and phylogeographic studies highly structured single, widespread species throughout central (e.g., Esselstyn and Brown, 2009; Brown et al., 2010; Linkem and western Mindanao, although conservative, presents some obvi- et al., 2010a, 2010b; Siler et al., 2010, 2011) suggest that within-is- ous and immediate conservation considerations (see below). land evolutionary diversification may contribute substantially more to Philippine megadiversity than previously recognized 4.2. Phylogeography, geographic distribution of genetic variation, and (Brown et al., 2008; Brown and Diesmos, 2009; Brown and Stuart, demography in press). If intra-island landscape processes have contributed sub- stantially to vertebrate diversification in other taxa as they have Our widespread geographic sampling across the range of Philip- influenced diversification in Ansonia, an expanded characterization pine Ansonia represents the first densely sampled phylogeographic of the evolutionary processes of diversification leading to study for an endemic Mindanao Island vertebrate. Many previous Philippine megadiversity may be required (Brown et al., 2002; works have surmised the existence of montane subcenters of bio- Brown and Diesmos, 2002, 2009; Esselstyn et al., 2010). diversity, or pockets of species endemism on this large, geologi- Our preferred phylogenetic estimate (Fig. 1) suggests a general cally complex amalgamation of paleoislands (Taylor, 1920, west-to-east dispersal into central and eastern Mindanao. We 1922a, 1928; Inger, 1954; Leviton, 1963; Kennedy et al., 2000; have, however, rejected a strict, west-to-east stepping-stone model Brown et al., 2002; Brown and Diesmos, 2009; Heaney et al., of dispersion or expansion along the linear Zamboanga Peninsula 2006a, 2006b, 2010). However, no previous studies have used ro- (Fig. 3a), and several intuitive dispersal-corridor hypotheses are bust genetic sampling to survey for Areas of Endemism (AOEs; similarly rejected (Fig. 3a–e). Evans et al., 2003b) across the Mindanao faunal region, nor as- Exploration of the approximate timing of diversification (see sessed intraspecific genetic partitioning within a widespread below) suggests that Ansonia dispersed simultaneously from endemic Mindanao species. west-central to east-central, southern, and eastern Mindanao, Our results unequivocally demonstrate for the first time that and that stable populations were established (possibly by con- widespread endemic Mindanao vertebrates exhibit significant geo- straints on dispersal imposed by larval biology) in the Pleistocene. graphical population structure and allopatric genetic AOEs, coinci- Given the past several million years of continued uplift of central dent with all the major upland geological components of the island Mindanao, and numerous oscillations of sea level and terrestrial (Peterson and Heaney, 1993; Heaney et al., 2005, 2006a, 2006b; climate during the Pleistocene (review: Woodruff, 2010), we sus- Esselstyn et al., 2009; Siler et al., 2011). The implications for ende- pect that today’s regionalized, geographically structured A. muelleri mic Mindanao vertebrates known from mid- to high-elevation for- populations result from Pleistocene habitat fragmentation driven ested regions of the island may be profound. If many of Mindanao’s by climate change (i.e., Evans et al., 2003b, 2008). Evidence sug- montane vertebrate species exhibit similar, highly divergent lin- gests that interglacial periods were (and are today) characterized eages (i.e., close to or bypassing species boundaries), not only have by rising sea levels, increased aridification of low-lying valleys ESUs for conservation gone largely undetected but species-level and wide inter-montane plains near sea level (Whitmore, 1984, biodiversity on the whole may be severely underestimated (Brown 1987; Lisiecki and Raymo, 2005; Bintanja et al., 2005; Corlett, and Diesmos, 2009). 2009; Woodruff, 2010; Wurster et al., 2010), suggesting expansion 344 M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350 of environmental barriers to dispersal for anurans with torrent- (2.7–3.0% uncorrected p-distance) and the inference of an 20 specialized larvae. A general concordance between the level of my age of the Mindanao clade, we note that an unrealistically divergence between the geographically structured clades of A. low nucleotide substitution rate (on the order of 0.000000135– muelleri (1.6–3.8%; Table 3), the timing of diversification esti- 0.00000015%; far slower than ever empirically recorded in mated for nodes within A. muelleri (Fig. 6), the approximate age mitochondrial gene sequences of amphibians) is implied by of onset of the decline in effective population size inferred by the Matsui et al.’s (2010) result. As a final point, we note that if our Bayesian skyline analysis (Fig. 5), and the known timing of Pleisto- approximate divergence dating estimation is reasonably accurate, cene sea-level and climatic oscillations (Lisiecki and Raymo, 2005; the timing of diversification inferred here is generally consistent Woodruff, 2010) somewhat corroborates our findings and with the Pleistocene sea-level oscillation (and climate-habitat frag- strengthens our confidence in our interpretation. mentation) model of diversification previously postulated (Heaney, Coupled with much lower levels of haplotype diversity, poly- 1985; Heaney et al., 2005; Esselstyn and Brown, 2009; Linkem morphic sites, and nucleotide diversity in A. mcgregori as com- et al., 2010a, 2010b; Siler et al., 2010) as a common mechanism pared to A. muelleri (Table 2), plus a significant Fu’s Fs in the that may have promoted vertebrate species diversification in the former but not the latter (Table 4), our results suggest a depar- archipelago (Brown and Diesmos, 2002, 2009). ture from expectations based on a null model of constant popula- tion size for A. mcgregori. Additionally, mismatch distributions for 4.4. Conservation implications this range-restricted species are smooth and unimodal, and the nonsignificant T statistic fails to reject the hypothesis of an Currently both Philippine species of Ansonia are listed in the expanding population in A. mcgregori. On the other hand, the rag- IUCN Red List of Threatened Species conservation status definitions ged, multimodal mismatch distribution and statistically signifi- (IUCN, 2010) as ‘‘Vulnerable (B1ab(iii) ‘‘populations trend: cant T for A. muelleri, suggest a stable, structured population. decreasing’’). Given our elucidation of A. mcgregori as a potentially

Finally, Ramos-Onsins and Rozas R2 statistics were small and sig- range-restricted species, we favor the continuation of this classifi- nificant in both species, suggesting recent demographic expansion cation. However, we note that no data have ever been presented to in some parts of their ranges. The signal of recent range expan- suggest that population sizes are declining in A. mcgregori. sion in A. mcgregori may be influenced by the fact that we have In contrast our analysis has identified in A. muelleri five diver- collected it in Pasonanca at the tip of the Zamboanga Peninsula, gent mitochondrial gene lineages, which may be considered con- and also at Murias (midway up the peninsula), but not at any servation ESUs. To preserve genetic diversity as a much-desirable intervening or other sites. component of biodiversity (Moritz, 1994; Evans et al., 2003b; Row- ley et al., 2009), protection of habitat in all five areas of genetic 4.3. Timing of diversification endemism across the range of A. muelleri will be required. We sus- pect that deep divergences within A. muelleri (Fig. 6) may warrant Our Bayesian relaxed clock analyses (Fig. 6) yielded broadly taxonomic partitioning (i.e., additional species recognition) overlapping highest posterior density intervals at basal nodes in pending the identification of diagnostic characters of morphology the tree, suggesting a similar temporal framework (the Pleisto- and/or bioacoustics. cene) for the divergences that produced the major structure, giving Two lineages revealed here require immediate taxonomic rise to the four putative species in our analyses. Our results suggest and conservation status assessments. A highly genetically diver- that species diversity within Mindanao Ansonia probably has gent lineage (‘‘Ansonia sp. 1’’), forming the sister taxon to all evolved within the last two million years. Interestingly, inferred remaining Philippine Ansonia was sampled at two mid-peninsular divergences within A. muelleri are approximately the same or even Zamboanga sites: Murias and Timolan. This morphologically slightly older than divergences inferred between A. sp. 1, A. sp. 2, indistinguishable lineage was encountered sympatrically and syn- and A. mcgregori (1.6–1.5 mya; Fig. 6), further emphasizing the topically with A. mcgregori at these two sites and may represent a potential for future taxonomic partitioning of A. muelleri and a gen- new, undescribed, ‘‘cryptic’’ species. Similarly, two unvouchered eral Pleistocene framework for diversification of Mindanao genetic samples taken from individuals at the Lake Lanao popula- Ansonia. tion (‘‘Ansonia sp. 2’’) represent a highly divergent lineage placed We consider Matsui et al.’s (2010) 20 my estimate of the age as the sister taxon to A. mcgregori + A. muelleri. Our suspicion is of Philippine Ansonia suspect (biased towards unrealistically old that both of these (and, conceivably, other as of yet undiscovered estimates of clade age) for three main reasons. First, the geological lineages) may represent range-restricted, unrecognized species in components of Mindanao were widely separated and largely still need of immediate conservation planning and/or taxonomic rec- submerged 20 mya (Hall, 1996, 1998; Yumul et al., 2003, 2004, ognition. Like the five divergent lineages of A. muelleri, these 2008, 2009), casting doubt on the possibility that vertebrates of two highly divergent western Mindanao lineages should be trea- low relative dispersal abilities could achieve long-distance dis- ted as ESUs for conservation purposes and should be scrutinized persal necessary to invade the separate paleoisland precursors of further for diagnostic differences in adult and larval morphology today’s Mindanao. Second, the Matsui et al. (2010) analysis or acoustic characters that may facilitate their recognition as dis- inferred a mean Ansonia–Pelophryne split of 74.9 (95% CI: 56.4– tinct species. 95.2) Mya and an age of 66.4 (49.5–84.0) Mya for the genus Anso- Finally, future phylogeographic and conservation genetic stud- nia. In well-calibrated analyses involving multiple unlinked loci ies of Mindanao’s diverse fauna should target genetic sampling and multiple nuclear genes, these ages are roughly equivalent to from within the AOEs defined by our elucidation of geographi- the estimated age of the clade Hyloidea and, as such, are far older cally structured genetic variation in Ansonia. We would not be than any major divergences within Bufonidae that gave rise to surprised if these isolated montane components of Mindanao har- Asian bufonid genera (e.g., 25–40 mya; Roelants et al., 2007); bor many additional undiscovered species and undiagnosed ESUs thus, the Matsui et al. (2010) analysis appears unrealistically in other vertebrate groups. A multi-taxon comparative approach biased toward old ages across their tree (possibly an artifact of (e.g., Evans et al., 2003a, 2008; Brown et al., 2010; Setiadi et al., their external calibration procedure, extreme rate variation across in press) would be a particularly compelling approach for future the tree, or a combination of these and other factors; Graur and studies geared towards understanding the origins and mainte- Martin, 2004). Finally given the shallow divergences detected in nance of high endemic species diversity in the southern Mindanao populations included in Matsui et al.’s (2011) analysis Philippines. M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350 345

Acknowledgments the University of Kansas, Fulbright and Fulbright-Hayes Fellowships, and NSF DEB 0804115 to C.D.S., and NSF EF-0334952 and DEB We thank the Protected Areas and Wildlife Bureau (PAWB) of 0743491 to R.M.B. For the loans of specimens and/or access to mu- the Philippine Department of Environment and Natural Resources seum collections (museum abbreviations follow Leviton et al., (DENR; especially DENR Region 10 personnel) for facilitating col- 1985), we thank A. Campbell (KU), J. Vindum, R. Drewes, A. Leviton lecting and export permits necessary for this and related studies. (CAS), V.S. Palpal-latoc (PNM), J. Ferner, H. Mays (CMNH) and B. Clark At PAWB we are particularly grateful for the supportive efforts of (BM). Critical reviews of the manuscript were provided by J. Essels- T.M. Lim, A. Tagtag, C. Custodio, and J.L. De Leon. The Philippine tyn, B. Evans, and D. Blackburn. Finally, MBS would like to thank Commission on Higher Education (CHED) and Fr. Saturnino Urios Mindanao State University–Iligan Institute of Technology, J.C.U. University provided financial support for fieldwork for M.B.S. Young and J.Y.T. Sanguila for logistical, advice, guidance and support. Laboratory work was supported by CHED-PGMASEGS funds to M.B.S. and NSF DEB 0743491 and 0640737 to R.M.B. Appendix A Support for field work by R.M.B., C.D.S., and A.C.D. was provided by the Panorama Fund grants from The University of Kansas Natural Species identity collector number, museum catalog numbers, History Museum and Biodiversity Institute, a Self Fellowship from and localities for all samples included in this study.

Taxon Field No. Catalog No. General Area Specific collection locality Ansonia sp. 1 MBS 027 KU 323467 West Mindanao Mt. Murias, Murias River, Barangay Lourdes, Pagadian City, Zamboanga del Sur Province Ansonia sp. 1 MBS 031 KU 323471 West Mindanao Mt. Timolan, Barangay Limas, Municipality of Guipos, Pagadian City, Zamboanga del Sur Province Ansonia sp. 1 MBS 030 KU 323470 West Mindanao Mt. Timolan, Barangay Limas, Municipality of Guipos, Pagadian City, Zamboanga del Sur Province Ansonia sp. 1 MBS 026 KU 323466 West Mindanao Mt. Murias, Murias River, Barangay Lourdes, Pagadian City, Zamboanga del Sur Province Ansonia sp. 1 MBS 029 KU 323469 West Mindanao Mt. Timolan, Barangay Limas, Municipality of Guipos, Pagadian City, Zamboanga del Sur Province Ansonia sp. 2 ACD 3600 No voucher North (Central) Mindanao Near Lake Lanao, Marawi City, Lanao del Sur Province Ansonia sp. 2 ACD 3601 No voucher North (Central) Mindanao Near Lake Lanao, Marawi City, Lanao del Sur Province Ansonia mcgregori RMB 11079 KU 321113 West Mindanao Barangay La Paz, Sitio Nancy, Municipality of Pasonanca, Zamboanga City Ansonia mcgregori RMB 11273 KU 321119 West Mindanao Barangay La Paz, Sitio Nancy, Municipality of Pasonanca, Zamboanga City Ansonia mcgregori RMB 10432 KU 314231 West Mindanao Barangay Baluno, Municipality of Pasonanca, Zamboanga City Ansonia mcgregori MBS 008 KU 323448 West Mindanao Mt. Murias, Murias River, Barangay Lourdes, Municipality of Pagadian City, Zamboanga del Sur Province Ansonia mcgregori MBS 005 KU 323445 West Mindanao Mt. Murias, Murias River, Barangay Lourdes, Municipality of Pagadian City, Zamboanga del Sur Province Ansonia mcgregori MBS 010 KU 323450 West Mindanao Murias River, Barangay Lourdes, Mt. Murias, Pagadian City, Zamboanga del Sur Ansonia mcgregori MBS 012 KU 323452 West Mindanao Murias River, Barangay Lourdes, Mt. Murias, Pagadian City, Zamboanga del Sur Ansonia mcgregori RMB 10395 KU 314194 West Mindanao Barangay Baluno, Municipality of Pasonanca, Zamboanga City Ansonia mcgregori RMB 10371 KU 314180 West Mindanao Barangay Baluno, Municipality of Pasonanca, Zamboanga City Ansonia mcgregori RMB 11003 KU 321106 West Mindanao Barangay Baluno, Municipality of Pasonanca, Zamboanga City Ansonia mcgregori MBS 011 KU 323451 West Mindanao Mt. Murias, Murias River, Barangay Lourdes, Municipality of Pagadian City, Zamboanga del Sur Province Ansonia mcgregori MBS 020 KU 323460 West Mindanao Mt. Murias, Murias River, Barangay Lourdes, Municipality of Pagadian City, Zamboanga del Sur Province Ansonia mcgregori MBS 009 KU 323451 West Mindanao Mt. Murias, Murias River, Barangay Lourdes, Municipality of Pagadian City, Zamboanga del Sur Province Ansonia mcgregori MBS 023 KU 323463 West Mindanao Mt. Murias, Murias River, Barangay Lourdes, Municipality of Pagadian City, Zamboanga del Sur Province Ansonia mcgregori MBS 016 KU 323456 West Mindanao Mt. Murias, Murias River, Barangay Lourdes, Municipality of Pagadian City, Zamboanga del Sur Province Ansonia mcgregori MBS 007 KU 323447 West Mindanao Mt. Murias, Murias River, Barangay Lourdes, Municipality of Pagadian City, Zamboanga del Sur Province Ansonia mcgregori MBS 018 KU 323458 West Mindanao Mt. Murias, Murias River, Barangay Lourdes, Municipality of Pagadian City, Zamboanga del Sur Province Ansonia mcgregori MBS 021 KU 323461 West Mindanao Mt. Murias, Murias River, Barangay Lourdes, Municipality of Pagadian City, Zamboanga del Sur Province Ansonia mcgregori MBS 019 KU 323459 West Mindanao Mt. Murias, Murias River, Barangay Lourdes, Municipality of Pagadian City, Zamboanga del Sur Province Ansonia mcgregori MBS 022 KU 323462 West Mindanao Mt. Murias, Murias River, Barangay Lourdes, Municipality of Pagadian City, Zamboanga del Sur Province Ansonia muelleri MBS 044 KU 323484 North (Central) Mindanao Near Mt. Gabunan, Sitio Minsaliding, Barangay Rogongon, Municipality of Iligan City, Lanao del Norte Province Ansonia muelleri MBS 035 KU 323475 North (Central) Mindanao Near Mt. Gabunan, Sitio Minsaliding, Barangay Rogongon, Municipality of Iligan City, Lanao del Norte Province Ansonia muelleri MBS 041 KU 323481 North (Central) Mindanao Near Mt. Gabunan, Sitio Minsaliding, Barangay Rogongon, Municipality of Iligan City, Lanao del Norte Province

(continued on next page) 346 M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350

Appendix A (continued)

Taxon Field No. Catalog No. General Area Specific collection locality Ansonia muelleri MBS 043 KU 323483 North (Central) Mindanao Near Mt. Gabunan, Sitio Minsaliding, Barangay Rogongon, Municipality of Iligan City, Lanao del Norte Province Ansonia muelleri ACD 3543 PNM⁄ North (Central) Mindanao Mt. Kitanglad, Bukidnon Province Ansonia muelleri MBS 036 KU 323476 North (Central) Mindanao Near Mt. Gabunan, Sitio Minsaliding, Barangay Rogongon, Municipality of Iligan City, Lanao del Norte Province Ansonia muelleri MBS 046 KU 323486 North (Central) Mindanao Near Mt. Gabunan, Sitio Minsaliding, Barangay Rogongon, Municipality of Iligan City, Lanao del Norte Province Ansonia muelleri RMB 643 CMNH⁄ Apo Massif Mt. Apo, Barangay Baracatan, Municipality of Toril, Davao del sur Province Ansonia muelleri RMB 639 CMNH⁄ Apo Massif Mt. Apo, Barangay Baracatan, Municipality of Toril, Davao del sur Province Ansonia muelleri RMB 640 CMNH⁄ Apo Massif Mt. Apo, Barangay Baracatan, Municipality of Toril, Davao del sur Province Ansonia muelleri RMB 641 CMNH⁄ Apo Massif Mt. Apo, Barangay Baracatan, Municipality of Toril, Davao del sur Province Ansonia muelleri RMB 639 CMNH⁄ Apo Massif Mt. Apo, Barangay Baracatan, Municipality of Toril, Davao del sur Province Ansonia muelleri RMB 642 CMNH⁄ Apo Massif Mt. Apo, Barangay Baracatan, Municipality of Toril, Davao del sur Province Ansonia muelleri ACD 1617 CMNH⁄ Apo Massif Mt. Apo, Barangay Baracatan, Municipality of Toril, Davao del sur Province Ansonia muelleri ACD 1640 CMNH⁄ Apo Massif Mt. Apo, Barangay Baracatan, Municipality of Toril, Davao del sur Province Ansonia muelleri ACD 4265 KU 319745 North (Central) Mindanao Mt. Balatukan, Barangay Lunotan, Sitio San Isidro, Municipality of Gingoog City, Misamis Oriental Province Ansonia muelleri ACD 4437 KU 319747 North (Central) Mindanao Mt. Balatukan, Barangay Lunotan, Sitio San Isidro, Municipality of Gingoog City, Misamis Oriental Province Ansonia muelleri ACD 4421 KU 319752 North (Central) Mindanao Mt. Balatukan, Barangay Lunotan, Sitio San Isidro, Municipality of Gingoog City, Misamis Oriental Province Ansonia muelleri ACD 4284 KU 319755 North (Central) Mindanao Mt. Balatukan, Barangay Lunotan, Sitio San Isidro, Municipality of Gingoog City, Misamis Oriental Province Ansonia muelleri ACD 4282 KU 319725 North (Central) Mindanao Mt. Balatukan, Barangay Lunotan, Sitio San Isidro, Municipality of Gingoog City, Misamis Oriental Province Ansonia muelleri ACD 4260 KU 319733 North (Central) Mindanao Mt. Balatukan, Barangay Lunotan, Sitio San Isidro, Municipality of Gingoog City, Misamis Oriental Province Ansonia muelleri ACD 4264 KU 319762 North (Central) Mindanao Mt. Balatukan, Barangay Lunotan, Sitio San Isidro, Municipality of Gingoog City, Misamis Oriental Province Ansonia muelleri ACD 4263 KU 319758 North (Central) Mindanao Mt. Balatukan, Barangay Lunotan, Sitio San Isidro, Municipality of Gingoog City, Misamis Oriental Province Ansonia muelleri ACD 4334 KU 319721 North (Central) Mindanao Mt. Balatukan, Barangay Lunotan, Sitio San Isidro, Municipality of Gingoog City, Misamis Oriental Province Ansonia muelleri MBS 064 KU 323504 North (Central) Mindanao Cagang-awan River, Sitio Ananaso, Barangay Dumalaguing, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 076 KU 323516 North (Central) Mindanao Cagang-awan River, Sitio Ananaso, Barangay Dumalaguing, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 047 KU 323487 North (Central) Mindanao Yandang, km. 30, Calabugao River, Sitio Nasandigan, Barangay Calabugao, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 075 KU 323515 North (Central) Mindanao Cagang-awan River, Sitio Ananaso, Barangay Dumalaguing, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 073 KU 323513 North (Central) Mindanao Cagang-awan River, Sitio Ananaso, Barangay Dumalaguing, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 074 KU 323514 North (Central) Mindanao Cagang-awan River, Sitio Ananaso, Barangay Dumalaguing, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 066 KU 323506 North (Central) Mindanao Cagang-awan River, Sitio Ananaso, Barangay Dumalaguing, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 077 KU 323517 North (Central) Mindanao Cagang-awan River, Sitio Ananaso, Barangay Dumalaguing, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 067 KU 323507 North (Central) Mindanao Cagang-awan River, Sitio Ananaso, Barangay Dumalaguing, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 052 KU 323492 North (Central) Mindanao Yandang, km. 30, Calabugao River, Sitio Nasandigan, Barangay Calabugao, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 056 KU 323496 North (Central) Mindanao Yandang, km. 30, Calabugao River, Sitio Nasandigan, Barangay Calabugao, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 057 KU 323497 North (Central) Mindanao Yandang, km. 30, Calabugao River, Sitio Nasandigan, Barangay Calabugao, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 051 KU 323491 North (Central) Mindanao Yandang, km. 30, Calabugao River, Sitio Nasandigan, Barangay Calabugao, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 049 KU 323489 North (Central) Mindanao Yandang, km. 30, Calabugao River, Sitio Nasandigan, Barangay Calabugao, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 054 KU 323494 North (Central) Mindanao Yandang, km. 30, Calabugao River, Sitio Nasandigan, Barangay Calabugao, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 050 KU 323490 North (Central) Mindanao Yandang, km. 30, Calabugao River, Sitio Nasandigan, Barangay Calabugao, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 053 KU 323493 North (Central) Mindanao Yandang, km. 30, Calabugao River, Sitio Nasandigan, Barangay Calabugao, Municipality of Impasugong, Bukidnon Province Ansonia muelleri MBS 095 KU 323520 South Mindanao Nobol River, Sitio Kapalanan, Barangay Gasi, Municipality of Kiamba, Sarangani Province Ansonia muelleri MBS 106 KU 323531 South Mindanao Nobol River, Sitio Kapalanan, Barangay Gasi, Municipality of Kiamba, Sarangani Province Ansonia muelleri MBS 100 KU 323525 South Mindanao Nobol River, Sitio Kapalanan, Barangay Gasi, M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350 347

Appendix A (continued)

Taxon Field No. Catalog No. General Area Specific collection locality Municipality of Kiamba, Sarangani Province Ansonia muelleri MBS 105 KU 323530 South Mindanao Nobol River, Sitio Kapalanan, Barangay Gasi, Municipality of Kiamba, Sarangani Province Ansonia muelleri MBS 107 KU 323532 South Mindanao Nobol River, Sitio Kapalanan, Barangay Gasi, Municipality of Kiamba, Sarangani Province Ansonia muelleri MBS 098 KU 323523 South Mindanao Nobol River, Sitio Kapalanan, Barangay Gasi, Municipality of Kiamba, Sarangani Province Ansonia muelleri MBS 094 KU 323519 South Mindanao Nobol River, Sitio Kapalanan, Barangay Gasi, Municipality of Kiamba, Sarangani Province Ansonia muelleri PNM/ PNM⁄ South Mindanao Mt. Busa, Municipality of Kiamba, CMNH 1718 Sarangani Province Ansonia muelleri PNM/ PNM⁄ South Mindanao Mt. Busa, Municipality of Kiamba, CMNH 1475 Sarangani Province Ansonia muelleri PNM/ PNM⁄ South Mindanao Mt. Busa, Municipality of Kiamba, CMNH 1472 Sarangani Province Ansonia muelleri PNM/ PNM⁄ South Mindanao Mt. Busa, Municipality of Kiamba, CMNH 1471 Sarangani Province Ansonia muelleri PNM/ PNM⁄ South Mindanao Mt. Busa, Municipality of Kiamba, CMNH 1643 Sarangani Province Ansonia muelleri PNM/ CMC 12173 South Mindanao Mt. Busa, Municipality of Kiamba, CMNH 1714 Sarangani Province Ansonia muelleri PNM/ CMC 12712 South Mindanao Mt. Busa, Municipality of Kiamba, CMNH 1644 Sarangani Province Ansonia muelleri PNM/ PNM⁄ South Mindanao Mt. Busa, Municipality of Kiamba, CMNH 1713 Sarangani Province Ansonia muelleri PNM/ CMC 12174 South Mindanao Mt. Busa, Municipality of Kiamba, CMNH 1756 Sarangani Province Ansonia muelleri PNM/ CMC 12176 South Mindanao Mt. Busa, Municipality of Kiamba, CMNH 1757 Sarangani Province Ansonia muelleri MBS 084 No voucher East Mindanao Bato River, Mt. Talinis – Bato, Barangay Cabugo, Municipality of Claver, Surigao del Norte Province Ansonia muelleri MBS 088 No voucher East Mindanao Bato River, Mt. Talinis – Bato, Barangay Cabugo, Municipality of Claver, Surigao del Norte Province Ansonia muelleri MBS 081 No voucher East Mindanao Bato River, Mt. Talinis – Bato, Barangay Cabugo, Municipality of Claver, Surigao del Norte Province Ansonia muelleri MBS 082 No voucher East Mindanao Bato River, Mt. Talinis – Bato, Barangay Cabugo, Municipality of Claver, Surigao del Norte Province Ansonia muelleri EMD 308 PNM⁄ East Mindanao Barangay San Antonio, Municipality of Remedios T. Romualdez, Agusan del Norte Province Ansonia muelleri MBS 002 KU 323442 East Mindanao Agay River, Mt. Bato – Batohon (in the Mt. Hilong – Hilong Range), Barangay San Antonio, Municipality of Remedios T. Romualdez, Agusan del Norte Province Ansonia muelleri MBS 001 KU 323441 East Mindanao Agay River, Mt. Bato – Batohon (in the Mt. Hilong – Hilong Range), Barangay San Antonio, Municipality of Remedios T. Romualdez, Agusan del Norte Province Ansonia muelleri EMD 301 PNM⁄ East Mindanao Agay River, Mt. Bato – Batohon (in the Mt. Hilong – Hilong Range), Barangay San Antonio, Municipality of Remedios T. Romualdez, Agusan del Norte Province Ansonia muelleri EMD 302 PNM⁄ East Mindanao Agay River, Mt. Bato – Batohon (in the Mt. Hilong – Hilong Range), Barangay San Antonio, Municipality of Remedios T. Romualdez, Agusan del Norte Province Ansonia muelleri ACD 3836 KU 319522 East Mindanao Mt. Magdiwata, Barangay Bayugan 2, Municipality of San Francisco, Agusan del Sur Province Ansonia muelleri ACD 3871 KU 319527 East Mindanao Mt. Magdiwata, Barangay Bayugan 2, Municipality of San Francisco, Agusan del Sur Province Ansonia muelleri ACD 3958 KU 319522 East Mindanao Mt. Magdiwata, Barangay Bayugan 2, Municipality of San Francisco, Agusan del Sur Province Ansonia muelleri GGT 043 PNM⁄ East Mindanao Purok Kulapi, Barangay Bahi, Municipality of Maragusan, Compostela Valley, Davao Province Ansonia muelleri GGT 051 PNM⁄ East Mindanao Purok Kulapi, Barangay Bahi, Municipality of Maragusan, Compostela Valley, Davao Province Ansonia muelleri GGT 065 PNM⁄ East Mindanao Purok Kulapi, Barangay Bahi, Municipality of Maragusan, Compostela Valley, Davao Province Ansonia muelleri ACD 2631 PNM⁄ East Mindanao Mt. Hamiguitan, Davao Oriental Province Ansonia muelleri ACD 2623 PNM⁄ East Mindanao Mt. Hamiguitan, Davao Oriental Province Ansonia muelleri ACD 2638 PNM⁄ East Mindanao Mt. Hamiguitan, Davao Oriental Province Ansonia muelleri ACD 2669 PNM⁄ East Mindanao Mt. Hamiguitan, Davao Oriental Province Ansonia muelleri ACD 2702 PNM⁄ East Mindanao Mt. Hamiguitan, Davao Oriental Province Ansonia muelleri ACD 2639 KU 326765 East Mindanao Mt. Hamiguitan, Davao Oriental Province

References Bickford, D., Lohman, D.J., Sodhi, N.S., Ng, P.K.L., Meier, R., Winker, K., Ingram, K.K., Das, I., 2007. Cryptic species as a window on diversity and conservation. Trends Ecol. Evol. 22, 148–155. Alcala, A.C., 1986. Guide to Philippine Flora and Fauna, vol. X, Amphibians and Bintanja, R., R van de Wal, S.W., Oerlemans, J., 2005. Modeled atmospheric Reptiles. Natural Resource Management Center, Ministry of Natural Resource tempera- Management Center, Ministry of Natural Resources and the University of the tures and global sea levels over the past million years. Nature 437, 125–128. Philippines, Manila, Philippines. Boulenger, G.A., 1882. Catalogue of the Batrachia Saliente s Ecaudata in the Alcala, A.C., Brown, W.C., 1998. Philippine Amphibians: an Illustrated Field Guide. Collection of the British Museum. Taylor and Francis, London, England. Bookmark Press, Makati City, Philippines. Boulenger, G.A., 1887. Descriptions of new reptiles and batrachians in the Amoroso, V.B., 2000. Status, species richness and ecosystem diversity in Mindanao British Museum (Natural History), Part III. Ann. Mag. Nat. Hist. Ser. 20, Island. In Anonymous (Ed.), Saving the Hottest of the Hotspots. Proceedings of the 50–53. National Biodiversity Conservation Priority-Setting Workshop, Mindanao Regional Boulenger, G.A., 1920. A monograph of the South Asian, Papuan, Melanesian and Consultation, Malagos Garden Resort, Davao City, Philippines, pp. 17–32. Australian frogs of the genus Rana. Records Indian Mus. 20, 1–126. 348 M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350

Brooks, T.M., Mittermeier, R.A., Mittermeier, C.G., da Fonseca, G.A.B., Rylands, A.B., Esselstyn, J.A., Timm, R.M., Brown, R.M., 2009. Do geological or climatic processes Konstant,W.R.,Flick,P.,Pilgrim,J.,Oldfield, S., Magin, G., Hilton-Taylor, C., 2002. drive speciation in dynamic archipelagos? The evolution of shrew diversity on Habitat loss and extinction in the hotspots of biodiversity. Conserv. Biol. 16, 909–923. SE Asian islands. Evolution 63, 2595–2610. Brown, R.M., 2006. The use of genetic data for inferring species limits. In: Catibog- Esselstyn, J.A., Oliveros, C.H., Moyle, R.G., Peterson, A.T., McGuire, J.A., Brown, R.M., Sinha, C.S., Heaney, L.R. (Eds.), Philippine Biodiversity: Principles and Practice. 2010. Integrating phylogenetic and taxonomic evidence illuminates complex Haribon Foundation for Conservation of Natural Resources, Quezon City, biogeographic patterns along Huxley’s modification of Wallace’s Line. J. Philippines, pp. 87–90. Biogeogr. 37, 2054–2066. Brown, R.M., 2007. Introduction to Robert F. Inger’s systematics and zoogeography Evans, B.J., Brown, R.M., McGuire, J.A., Supriatna, J., Andayani, N., Diesmos, A.C., of Philippine amphibia. In: Inger, R.F. (Ed.), Systematics and Zoogeography of Iskandar, D., Melnick, D.J., Cannatella, D.C., 2003a. Phylogenetics of fanged Philippine Amphibia. Natural History Publications, Kota Kinabalu, Malaysia, pp. frogs: testing biogeographical hypotheses at the interface of the Asian and 1–17. Australian faunal zones. Syst. Biol. 52, 794–819. Brown, W.C., Alcala, A.C., 1970. The zoogeography of the Philippine Islands, a Evans, B.J., Supriatna, J., Andayani, N., Setiad, M.I., Cannatella, D.C., Melnick, D.J., fringing archipelago. Proc. California Acad. Sci. 38, 105–130. 2003b. Monkeys and toads define areas of genetic endemism on the island of Brown, W.C., Alcala, A.C., 1978. Philippine Lizards of the Family Gekkonidae. Sulawesi. Evolution 57, 1436–1443. Silliman University Press, Dumaguete City, Philippines. Evans, B.J., McGuire, J.A., Brown, R.M., Andayani, N., Supriatna, J., 2008. A coalescent Brown, W.C., Alcala, A.C., 1980. Philippine Lizards of the Family Scincidae. Silliman framework for comparing alternative models of population structure with University Press, Dumaguete City, Philippines. genetic data: evolution of Celebes toads. Biol. Lett. 4, 430–433. Brown, W.C., Alcala, A.C., 1982. Modes of reproduction of Philippine anurans. In: Excoffier, L., Laval, G., Schneider, S., 2005. Arlequin ver. 3.0: An integrated software Rodin, A.G.J., Miyata, K. (Eds.), Advances in Herpetology and Evolutionary package for population genetics data analysis. Evol. Bioinf. Online 1, 47–50. Biology. Museum of Comparative Biology, Cambridge, MA, USA, pp. 416–428. Fu, Y.X., 1997. Statistical tests of neutrality of mutations against population growth, Brown, W.C., Alcala, A.C., 1994. Philippine frogs of the family Rhacophoridae. Proc. hitchhiking, and background selection. Genetics 147, 915–925. California Acad. Sci. 48, 185–220. Galbreath, K.E., Hafrner, D.J., Zamudio, K.R., 2009. When cold is better: climate- Brown, R.M., Diesmos, A.C., 2002. Application of lineage-based species concepts to driven elevation shifts yield complex patterns of diversification and oceanic island frog populations: the effects of differing taxonomic philosophies demography in an alpine specialist (American Pika, Ochotona princeps). on the estimation of Philippine biodiversity. Silliman J. 42, 133–162. Evolution 63, 2848–2863. Brown, R.M., Diesmos, A.C., 2009. Philippines, biology. In: Gillespie, R., Clague, D. Graur, D., Martin, W., 2004. Reading the entrails of chickens: molecular timescales (Eds.), Encyclopedia of Islands. University of California Press, Berkeley, pp. 723– of evolution and the illusion of precision. Trend. Genet. 20, 80–86. 732. Guindon, S., Gascuel, O., 2003. A simple, fast and accurate method to estimate large Brown, R.M., Guttman, S.I., 2002. Phylogenetic systematics of the Rana signata phylogenies by maximum-likelihood. Syst. Biol. 52, 696–704. complex of Philippine and Bornean stream frogs: reconsideration of Huxley’s Hall, R., 1996. Reconstructing Cenozoic SE Asia. In: Hall, R., Blundell, D. (Eds.), modification of Wallace’s Line at the Oriental–Australian faunal zone interface. Tectonic Evolution of Southeast Asia. Geological Society, London, England, pp. Biol. J. Linn. Soc. 76, 393–461. 153–184. Brown, R.M., Stuart, B.L., in press. Patterns of biodiversity discovery through time: Hall, R., 1997. Cenozoic plate tectonics reconstructions of SE Asia. In: Fraser, A.J., an historical analysis of defining amphibian species boundaries in the Southeast Mathews, S.J., Murphy, R.W. (Eds.), Petrolium Geology of Southeast Asia. Asian mainland and island archipelagos. Southeast Asian Gateway to Evolution Geolo- Proceedings Volume. Trustees of the the Natural History Museum, London. gical Society of London, Special Publication 126, London, England, pp. 11–24. Brown, R.M., McGuire, J.A., Diesmos, A.C., 2000. Status of some Philippine frogs Hall, R., 1998. The plate tectonics of Cenozoic SE Asia and the distribution of land related to Rana everetti (Anura: Ranidae), description of a new species, and and sea. In: Hall, R., Holloway, J.D. (Eds.), Biogeography and Geological resurrection of Rana igorota (Taylor 1922). Herpetologica 56, 81–104. Evolution of SE Asia. Backhuys Publishers, Leiden, Netherlands, pp. 99–131. Brown, R.M., Diesmos, A.C., Alcala, A.C., 2002. The state of Philippine herpetology Hall, R., 2002. Cenozoic geological and plate tectonic evolution of SE Asia and the and the challenges for the next decade. Silliman J. 42, 18–87. SW Pacific: computer-based reconstructions and animations. J. Asian Earth Sci. Brown, R.M., Diesmos, A.C., Alcala, A.C., 2008. Philippine amphibian biodiversity is 20, 353–434. increasing in leaps and bounds. In: Stuart, S.N., Hoffmann, M., Chanson, J.S., Cox, Hamilton, W., 1979. Tectonics of the Indonesian Region. United States Geological N.A., Berridge, R, Ramani, P., Young, B.E. (Eds.), Threatened Amphibians of the Survey Professional Papers 1078, pp. 1–345. World. Lynx Ediciones, Barcelona, Spain; IUCN – The World Conservation Union, Hare, M.P., 2001. Prospects for nuclear gene phylogeography. Trends Ecol. Evol. 16, Gland, Switzerland; and Conservation International, Arlington, pp. 82–83. 700–706. Brown, R.M., Siler, C.D., Diesmos, A.C., Alcala, A .C., 2009. Philippine frogs of the Harpending, H.C., Batzer, M.A., Gurven, M., Jorde, L.B., Sherry, S.T., 1998. Genetic genus Leptobrachium (Anura; Megophryidae): taxonomic revision, phylogeny- traces of ancient demography. Proc. Natl. Acad. Sci. USA 95, 1961–1967. based species delimitation, and descriptions of three new species. Herpetol. Heaney, L.R., 1985. Zoogeographic evidence for middle and late Pleistocene land Monogr. 23, 1–44. bridges to the Philippine Islands. Modern Quaternary Research in Southeast Brown, R.M., Linkem, C.W., Siler, C.D., Sukumaran, J., Esselstyn, J.A., Diesmos, A.C., Asia 9, 127–144. Iskandar, D.T., Bickford, D., Evans, B.J., McGuire, J.A., Grismer, L., Supriatna, J., Heaney, L.R., Dolar, M.L., Balete, D.S., Esselstyn, J.A., Rickart, E.A., Sedlock, J.L., 2010. Andayani, N., 2010. Phylogeography and historical demography of Polypedates Synopsis of Philippine Mammals. Field Museum of Natural History and the leucomystax in the islands of Indonesia and the Philippines: evidence for recent Philippine Department of Environment and Natural Resources, Protected Areas human-mediated range expansion? Mol. Phylogenet. Evol. 57, 598–619. and Wildlife Bureau. Corlett, R.T., 2009. The Ecology of Tropical East Asia. Oxford University Press, (accessed October 2010). Oxford, England. Heaney, L.R., Walsh Jr., J.S., Peterson, A.T., 2005. The roles of geological history and Crawford, A.J., 2003a. Huge populations and old species of Costa Rican and colonization abilities in genetic differentiation between mammalian Panamanian dirt frogs inferred from mitochondrial and nuclear gene sequences. populations in the Philippine archipelago. J. Biogeog. 32, 229–247. Mol. Ecol. 12, 2525–2540. Heaney, L.R., Tabaranza Jr., B.R., Balete, D.S., Rigertis, N., 2006a. Synopsis and Crawford, A.J., 2003b. Relative rates of nucleotide substitution in frogs. J. Mol. Evol. biogeography of the mammals of Camiguin Island. Fieldiana 106, 28–48. 57, 636–641. Heaney, L.R., Tabaranza Jr., B.R., Rickart, E.A., Balete, D.S., Ingle, N.R., 2006b. The Delima, E.M.M., Ates, F.B., Ibañez, J.C., 2006. Species composition and microhabitats mammals of Mt. Kitanglad Nature Park, Mindanao, Philippines. Fieldiana 112, of frogs within Arakan Valley Conservation Area, Cotabato, Mindanao Island, 1–63. Philippines. Banwa 3, 16–30. Hoogstral, H., 1951. Philippine zoological expedition 1946–1947: narrative and Delima, E.M.M., Diesmos, A.C., Ibañez, J.C., 2007. The herpetofaunal importance of itinerary. Fieldiana 33, 1–86. Mt. Hamiguitan Range, Mindanao Island, Philippines. Banwa 4, 27–40. Hudson, R.R., 1990. Gene genealogies and the coalescent process. Oxford Surv. Evol. Diamond, J.M., Gilpin, M.E., 1983. Biogeographic umbilici and the origin of the Biol. 7, 1–44. Philippine avifauna. Oikos 41, 307–321. Inger, R.F., 1954. Systematics and zoogeography of Philippine Amphibia. Fieldiana Drummond, A.J., Rambaut, A., 2006. BEAST v1.4. . Inger, R.F., 1960. A review of the Oriental toads of the genus Ansonia Stoliczka. Drummond, A.J., Rambaut, A., 2007. Bayesian evolutionary analysis by sampling Fieldiana 39, 473–503. trees. BMC Evol. Biol. 7, 214. Inger, R.F., 1966. The systematics and zoogeography of the Amphibia of Borneo. Drummond, A.J., Rambaut, A., Shapiro, B., Pybus, O.G., 2005. Bayeian coalescent Fieldiana 52, 1–402. inference of past population dynamics from molecular sequences. Mol. Biol. Inger, R.F., 1985. Tadpoles of the forested regions of Borneo. Fieldiana Evol. 22, 1185–1192. 26, 1–89. Duellman, W.E., Trueb, L., 1994. Biology of Amphibians. Johns Hopkins University Inger, R.F., 1992. Variation of apomorphic characters in stream-dwelling tadpoles of Press, Baltimore. the bufonid genus Ansonia (Amphibia: Anura). Zool. J. Linnean Soc. 105, 225– Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy and 237. high throughput. Nucl. Acids Res. 32, 1792–1797. Inger, R.F., 2005. The frog fauna of the Indo-Malayan region as it applies to Wallace’s Esselstyn, J.A., Brown, R.M., 2009. The role of repeated sea-level fluctuations in the line. In: Tuen, A.A., Das, I. (Eds.), Wallace in Sarawak – 150 Years Later: an generation of shrew (Soricidae: Crocidura) diversity in the Philippine International Conference of Biogeography and Biodiversity. Institute of Archipelago. Mol. Phylogenet. Evol. 53, 171–181. Biodiversity and Environmental Conservation, University Malaysia Sarawak, Esselstyn, J.A., Garcia, H.J.D., Saulog, M.G., Heaney, L.R., 2008. A new species of Kota Samarahan, Sarawak, pp. 82–90. Desmalopex (Pteropodidae) from the Philippines, with a phylogenetic analysis of IUCN 2010. IUCN Red List of Threatened Species. Version 2010.3. . M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350 349

Jones, A.W., Kennedy, R.S., 2008. Evolution in a tropical archipelago: comparative Ramos-Onsins, S.E., Rozas, J., 2002. Statistical properties of new neutrality tests phylogeography of Philippine fauna and flora reveals complex patterns of against population growth. Mol. Biol. Evol. 19, 2092–2100. colonization and diversification. Biol. J. Linn. Soc. 95, 620–639. Reid, W.V., 1998. Biodiversity hotspots. Trend. Ecol. Evol. 13, 275–280. Kennedy, R.S., Gonzales, P.C., Dickinson, E.C., Miranda, H.C., Fisher, T.H., 2000. A Roberts, T.E., Lanier, H.C., Sargis, E.J., Olson, L.E., 2011. Molecular phylogeny of Guide to the Birds of the Philippines. Oxford University Press, Oxford. treeshrews (Mammalia: Scandentia) and the timescale of diversification in Leviton, A.E., 1963. Remarks on the zoogeography of Philippine terrestrial snakes. Southeast Asia. Mol. Phylogenet. Evol., doi:10.1016/j.ympev.2011.04.021. Proc. California Acad. Sci. 42, 112–145. Roelants, K., Gower, D.J., Wilkinson, M., Loader, S.P., Biju, S.D., Guillaume, K., Moriau, Leviton, A.E., Gibbs Jr., R.H., Heal, E., Dawson, C.E., 1985. Standards in herpetology L., Bossuyt, F., 2007. Global patterns of diversification in the history of modern and ichthyology: Part I. Standard symbolic codes for institutional resource amphibians. Proc. Nat. Acad. Sci. 104, 887–892. collections in herpetology and ichthyology. Copeia 1985, 802–821. Rogers, A.R., Harpending, H., 1992. Population growth makes waves in the Lewis, S.D., 1997. Philippines. In: Moores, E.M., Fairbridge, R.W. (Eds.), Encyclopedia distribution of pairwise genetic differences. Mol. Biol. Evol. 9, 552–569. of European and Asian Regional Geology. Chapman and Hall, New York, pp. Ronquist, F., Huelsenbeck, J.P., 2003. MRBAYES 3: Bayesian phylogenetic inference 591–604. under mixed models. Bioinformatics 19, 1572–1574. Linkem, C.A., Diesmos, A.C., Brown, R.M., 2010a. A new scincid lizard Rowley, J., Brown, R., Bain, R., Kusrini, M., Inger, R.F., Stuart, B.L., Wogan, G., Chan- (Genus Sphenomorphus) from Palawan Island, Philippines. Herpetologica 66, ard, T., Trung, C.T., Diesmos, A.C., Iskandar, D.T., Lau, M., Ming, L.T., Makchai, S., 67–79. Thy, N., Truong, N.Q., Phimmachak, S., 2009. Impending conservation crisis for Linkem, C.A., Hesed, K., Diesmos, A.C., Brown, R.M., 2010b. Species boundaries and Southeast Asian amphibians. Biol. Lett. 6, 336–338. cryptic lineage diversity in a Philippine forest skink complex (Reptilia; Rozas, J., Sánchez-DelBarrio, J.C., Messeguer, X., Rozas, R., 2003. DNASP, DNA Squamata; Scincidae: Lygosominae). Mol. Phylogenet. Evol. 56, 572–585. polymorphism analyses by the coalescent and other methods. Bioinformatics Lisiecki, L.E., Raymo, M.E., 2005. A Pliocene–Pleistocene stack of 57 globally 19, 2496–2497. distributed benthic d18O records. Paleoceanography 20, PA1003. Setiadi, M.I., McGuire, J.A. Brown, R.M. Zubairi, M. Iskandar, D.T. Andayani, N., Macey, J.R., Schulte II, J.A., Larson, A., Fang, Z., Wang, Y., Tuniyev, B.S., Papenfuss, T.J., Supriatna, J., Evans, B.J., in press. Adaptive radiation and ecological opportunity 1998. Phylogenetic relationships of toads of the Bufo bufo species group from in Sulawesi fanged frogs (Limnonectes). Am. Nat. the eastern escarpment of the Tibetan Plateau: a case of vicariance and Shimodaira, H., 2002. An approximately unbiased test of phylogenetic tree dispersal. Mol. Biol. Evol 9, 80–87. selection. Syst. Biol. 51, 492–508. Macey, J.R., Strasburg, J.L., Brisson, J.A., Vredenburg, V.T., Jennings, M., Larson, A., Shimodaira, H., Hasegawa, M., 2001. CONSEL: for assessing the confidence of 2001. Molecular phylogenetics of western North American frogs of the Rana phylogenetic tree selection. Bioinformatics 17, 1246–1247. boylii species group. Mol. Phylogenet. Evol. 19, 131–143. Siler, C.D., Brown, R.M., 2010. Phylogeny-based species delimitation in Philippine Maddison, D.R., Maddison, W.P., 2005. MacClade: Analysis of Phylogeny and slender skinks (Reptilia: Squamata: Scincidae: Brachymeles): taxonomic Character Evolution. Vers. 4.08. Sinauer, Sunderland. revision of the pentadactyl species groups and description of three new Matsui, M., Nabhitabhata, J., Panha, S., 1998. A new Ansonia from Northern Thailand species. Herpetol. Monogr. 24, 1–54. (Anura: Bufonidae). Herpetologica 54, 448–454. Siler, C.D., Balete, D.S., Diesmos, A.C., Brown, R.M., 2010. A new legless loam- Matsui, M., Khonsue, W., Nabhitabhata, J., 2005. A new Ansonia from the Isthmus of swimming lizard (Reptilia: Squamata: Scincidae: Genus Brachymeles) from the Kra, Thailand (Amphibia, Anura, Bufonidae). Zool. Sci. 22, 809–814. Bicol Peninsula, Luzon Island, Philippines. Copeia 2010, 114–122. Matsui, M., Yambun, P., Sudin, A., 2007. Taxonomic relationships of Ansonia anotis Siler, C.D., Oaks, J.R., Esselstyn, J.A., Diesmos, A.C., Brown, R.M., 2010. Phylogeny and Inger, Tan, and Yambun, 2001 and Pedostibes maculatus (Mocquard, 1890), with biogeography of Philippine bent-toed geckos (Gekkonidae: Cyrtodactylus) a description of a new genus (Amphibia, Bufonidae). Zool. Sci. 24, 1159–1166. contradict a prevailing model of Pleistocene diversification. Mol. Phylogenet. Matsui, M., Tominaga, A., Liu, W., Khonsue, W., Grismer, L.L., Diesmos, A.C., Das, I., Evol. 55, 699–710. Sudin, A., Yambun, P., Yong, H., Brown, R.M., 2009. Phylogenetic Relationships of Siler, C.D., Diesmos, A.C., Alcala, A.C., Brown, R.M., 2011. Phylogeny of Philippine Ansonia from Southeast Asia inferred from mitochondrial DNA sequences: slender skinks (Scincidae: Brachymeles) reveals underestimated species systematic and biogeographic implications (Anura: Bufonidae). Mol. diversity, complex biogeographical relationships, and cryptic patterns of Phylogenet. Evol. 54, 561–570. lineage diversification. Mol. Phylogenet. Evol. 59, 53–65. Matsui, M., Tominaga, A., Liu, W., Khonsue, W., Grismer, L.L., Diesmos, A.C., Das, I., Slatkin, M., Hudson, R.R., 1992. Pairwise comparisons of mitochondrial DNA Sudin, A., Yambun, P., Yong, H., Sukumaran, J., Brown, R.M., 2010. Phylogenetic sequences in stable and exponential growth populations. Genetics 129, 555– relationships of Ansonia from Southeast Asia inferred from mitochondrial DNA 562. sequences: systematic and biogeographic implications (Anura: Bufonidae). Mol. Smith, M.A., 1930. The reptilia and amphibia of the Malay Peninsula. Bull. Raffles Phylogenet. Evol. 54, 561–570. Mus. 3, 1–149. McGuire, J.A., Alcala, A.C., 2000. A taxonomic revision of the flying lizards of the Smith, M.A., 1935. The Fauna of British India, including Ceylon and Burma. Reptilia Philippine Islands (Iguania: Agamidae: Draco), with a description of a new and Amphibia, II. Sauria. Taylor and Francis, London, England. species. Herpetol. Monogr. 14, 92–145. Smith, B.E., 1993a. Notes on a collection of squamate reptiles from eastern McGuire, J.A., Kiew, B.-H., 2001. Phylogenetic systematics of Southeast Asian flying Mindanao, Philippines Islands, Part 1: Lacertilia. Asiatic Herpetol. Res. 5, 85– lizards (Iguania: Agamidae:Draco) as inferred from mitochondrial DNA 95. sequence data. Biol. J. Linn. Soc. 72, 203–229. Smith, B.E., 1993b. Notes on a collection of squamate reptiles from eastern Mittermeier, R.A., P.Robles Gil, P.R., Mittermeier, G.G., 1997. Megadiversity: Earth’s Mindanao, Philippines Islands, Part 2: Serpentes. Asiatic Herpetol. Res. 5, 95– Biologically Wealthiest Nations. CEMEX, Monterrey, Mexico. 102. Mittermeier, R.A., Myers, N., Thomsen, J.B., da Fonseca, G.A.B., Olivieri, S., 1998. Stamatakis, A., 2006. RAxML-VI-HPC: maximum likelihood-based phylogenetic Biodiversity hotspots and major tropical wilderness areas: approaches to analyses with thousands of taxa and mixed models. Bioinformatics 22, 2688– setting conservation priorities. Conserv. Biol. 1, 516–520. 2690. Mittermeier, R.A., Myers, N., Gil, P.R., Mittermeier, G.G., 1999. Hotspots: Earth’s Stamatakis, A., Blagojevic, F., Nikolopoulos, D., Antonopoulos, C., 2007. Exploring Biologically Richest and Most Endangered Terrestrial Ecosystems. CEMEX, new search algorithms and hardware for phylogenetics: RAxML meets the IBM Monterrey, Mexico. cell. J. VLSI Signal Process. 48, 271–286. Moriarty, E.C., Cannatella, D.C., 2004. Phylogenetic relationships of North American Stamatakis, A., Hoover, P., Rougemont, J., 2008. A rapid bootstrap algorithm for the chorus frogs (Pseudacris). Mol. Phylogenet. Evol. 30, 409–420. RAxML web servers. Syst. Biol. 57, 758–771. Moritz, C., 1994. Defining ‘evolutionary significant units’ for conservation. Trend. Stuart, B.L., Bain, R.H., 2008. Amphibian species discovery in mainland Southeast Ecol. Evol. 9, 373–375. Asia. In: Stuart, S.N., Hoffmann, M., Chanson, J.S., Cox, N.A., Berridge, R.J., Nei, M., 1987. Molecular Evolutionary Genetics. Columbia University Press, New Ramani, P., Young, B.E. (Eds.), Threatened Amphibians of the World. Lynx York. Edicions, Barcelona, Spain; IUCN, Gland, Switzerland, and Conservation Nei, M., Tajima, F., 1981. DNA polymorphism detectable by restriction International, Arlington, pp. 83–84. endonuclease. Genetics 97, 145–163. Stuart, B.L., Inger, R.F., Voris, H.K., 2006. High level of cryptic species diversity Nichols, R., 2001. Gene trees and species trees are not the same. Trend. Ecol. Evol. revealed by sympatric lineages of Southeast Asian forest frogs. Biol. Lett. 2, 470– 16, 358–364. 474. Nuñeza, O.N., Ates, F.B., Alicante, A.A., 2010. Distribution of endemic and threatened Swofford, D.L., 2002. PAUP⁄. Phylogenetic Analysis Using Parsimony (⁄ and Other herpetofauna in Mt. Malindang, Mindanao, Philippines. Biodiv. Conserv. 19, Methods). Ver 4.0. Sinauer Associates, Sunderland. 503–518. Tan, A.M., Wake, D.B., 1995. MtDNA phylogeography of the California newt, Taricha Oliveros, C.H., Moyle, R.G., 2010. Origin and diversification of Philippine bulbuls. torosa (Caudata, Salimandridae). Mol. Phylogenet. Evol. 4, 383–394. Mol. Phylogenet. Evol. 54, 822–832. Taylor, E.H., 1915. New species of Philippine lizards. Philippine J. Sci. 10, 89–108. Peterson, A.T., Heaney, L.R., 1993. Genetic differentiation in Philippine bats of the Taylor, E.H., 1918. Reptiles of the Sulu archipelago. Philippine J. Sci. 13, 233–267. genera Cynopterus and Haplonycteris.. Biolog. J. Linnean Soc. 49, 203–218. Taylor, E.H., 1920. Philippine amphibia. Philippine J. Sci. 16, 213–359. Posada, D., in press. jModelTest: phylogenetic model averaging. Mol. Biol. Evol. Taylor, E.H., 1922a. The Lizards of the Philippine Islands. Philippine Bur. Sci., Pramuk, J.B., Robertson, T., Sites Jr., J.W., Noonan, B.P., 2007. Around the world in 10 Monogr. 17, Manila, Philippines. million years: biogeography of the nearly cosmopolitan true toads (Anura: Taylor, E.H., 1922b. The Snakes of the Philippine Islands. Department of Agriculture Bufonidae). Global Ecol. Biogeogr. 17, 72–83. and Natural Resources, Bureau of Science, Manila, Philippines. Quah, S.H.E., Grismer, L.L., Muin, M.A., Anuar, S.M.S., 2011. Re-discovery and re- Taylor, E.H., 1922c. Additions to the herpetological fauna of the Philippine Islands. I. description of Ansonia penangensis Stoliczka, 1870 (Anura: Bufonidae) from Philippine Journal of Science 21, 161–206. Penang Island, Malaysia. Zootaxa 2807, 57–64. Taylor, E.H., 1925. Additions to the herpetological fauna of the Philippine Islands, IV. Rambaut, A., Drummond, A.J., 2007. Tracer v1.4. . Philippine J. Sci. 26, 97–111. 350 M.B. Sanguila et al. / Molecular Phylogenetics and Evolution 61 (2011) 333–350

Taylor, E.H. 1928. Amphibians, lizards, and snakes of the Philippines. In: Dickerson, Wilgenbusch, J.C., Warren, D.L., Swofford, D.L., 2004. AWTY: a System for Graphical R. (Ed.), Distribution of Life in the Philippines. Philippine Bur. of Sci. Monogr. 21, Exploration of MCMC Convergence in Bayesian Phylogenetic Inference. . Taylor, E.H., 1975. Philippine adventures: an autobiographical memoir. In: Taylor, Woodruff, D.S., 2010. Biogeography and conservation in Southeast Asia: how 2.7 E.H., Leonard, A.B., Smith, H.M., Pisani, G.R. (Eds.), Recollections of an million years of repeated environmental fluctuations affect today’s patterns and Herpetologist. University of Kansas Museum of Natural History Monogr. 4, the future of the remaining refugial-phase biodiversity. Biodiver. Conserv. 19, Lawrence, pp. 1–105. 919–941. Van Kampen, P.N., 1923. The Amphibia of the Indo-Australian Archipelago. E.J. Brill Wurster, C.M., Bird, M.I., Bull, I.D., Creed, F., Bryant, C., Dungait, J.A.J., Paz, V., 2010. Ltd., Leiden. Forest contraction in north equatorial Southeast Asia during the Last Glacial Voris, H.K., 2000. Maps of Pleistocene sea levels in Southeast Asia: shorelines, river Period. Proc. Nat. Acad. Sci. 107, 15508–15511. systems and time durations. J. Biogeogr. 27, 1153–1167. Yumul Jr., G.P., Dimalanta, C.B., Tamayao Jr., R.A., Maury, R.C., 2003. Collision, Wang, I.J., Crawford, A.J., Bermingham, E., 2008. Phylogeography of the pygmy rain subduction and accretion events in the Philippines: a synthesis. Isl. Arc 12, 77– frog (Pristamantis ridens) across the lowland wet forests of isthmian Central 91. America. Mol. Phylogenet. Evol. 47, 992–1004. Yumul Jr., G.P., Dimalanta, C.B., Tamayao Jr., R.A., Maury, R.C., Bellon, H., Polvé, M., Welton, L.J., Siler, C.D., Diesmos, A.C., Brown, R.M., 2010a. Phylogeny-based species Maglambayan, V.B., Querubin, C.L., Cotton, J., 2004. Geology of the Zamboanga delimitation of southern Philippines bent-toed geckos and a new species of Peninsula, Philippines: an enigmatic South China Continental fragment? In: Cyrtodactylus (Squamata; Gekkonidae) from western Mindanao and the Sulu Malpas, J., Fletcher, C.J.N., Ali, J.R., Atchinson, J.C. (Eds.), Aspects of the Tectonic Archipelago. Zootaxa 2390, 49–68. Evolution of China, vol. 226. Geological Society of London, Special Publication, Welton, L.J., Siler, C.D., Diesmos, A.C., Linkem, C.W., Brown, R.M., 2010b. Philippine pp. 289–312. bent-toed geckos of the Cyrtodactylus agusanensis complex: multilocus Yumul Jr., G.P., Dimalanta, C.B., Maglambayan, V.B., 2008. Tectonic setting of a phylogeny, morphological diversity, and description of three new species. composite terrain: a review of the Philippine island arc system. Geosci. J. 12, Herpetol. Monogr. 24, 55–85. 7–17. Whitmore, T.C., 1984. Tropical Rain Forests of the Far East. Clarendon Press, Oxford, Yumul Jr., G., Dimilanta, C.B., Queaño, K., Marquez, E., 2009. Philippines, geology. In: England. Gillespie, R., Clague, D. (Eds.), Encyclopedia of Islands. University of California Whitmore, T.C., 1987. Biogeographical Evolution of the Malay Archipelago. Press, Berkeley, California, pp. 732–738. Clarendon Press, Oxford, England.